# DD Subagents — Full Listing Site: https://subagent.developersdigest.tech Generated: 2026-04-29T02:36:22.378Z Totals: 210 models, 20 frameworks, 25 MCP servers, 12 patterns, 17 skills. ## Models (210) ### Gemini 3.1 Pro Preview - URL: https://subagent.developersdigest.tech/models/gemini-3-1-pro-preview - Creator: Google - Intelligence score: 57 - Price ($/M tokens): 4.5 - Speed (tok/s): 119 - Latency (s): 34.87 - Context window: 1000000 ### GPT-5.4 (xhigh) - URL: https://subagent.developersdigest.tech/models/gpt-5-4 - Creator: OpenAI - Intelligence score: 57 - Price ($/M tokens): 5.63 - Speed (tok/s): 74 - Latency (s): 194.63 - Context window: 1000000 ### GPT-5.3 Codex (xhigh) - URL: https://subagent.developersdigest.tech/models/gpt-5-3-codex - Creator: OpenAI - Intelligence score: 54 - Price ($/M tokens): 4.81 - Speed (tok/s): 65 - Latency (s): 96.73 - Context window: 400000 ### Claude Opus 4.6 (max) - URL: https://subagent.developersdigest.tech/models/claude-opus-4-6-adaptive - Creator: Anthropic - Intelligence score: 53 - Price ($/M tokens): 10 - Speed (tok/s): 44 - Latency (s): 13.57 - Context window: 200000 ### Claude Sonnet 4.6 (max) - URL: https://subagent.developersdigest.tech/models/claude-sonnet-4-6-adaptive - Creator: Anthropic - Intelligence score: 52 - Price ($/M tokens): 6 - Speed (tok/s): 46 - Latency (s): 115.48 - Context window: 200000 ### GPT-5.2 (xhigh) - URL: https://subagent.developersdigest.tech/models/gpt-5-2 - Creator: OpenAI - Intelligence score: 51 - Price ($/M tokens): 4.81 - Speed (tok/s): 67 - Latency (s): 113.31 - Context window: 400000 ### GLM-5 - URL: https://subagent.developersdigest.tech/models/glm-5 - Creator: Z AI - Intelligence score: 50 - Price ($/M tokens): 1.55 - Speed (tok/s): 55 - Latency (s): 1.61 - Context window: 200000 ### GPT-5.2 Codex (xhigh) - URL: https://subagent.developersdigest.tech/models/gpt-5-2-codex - Creator: OpenAI - Intelligence score: 49 - Price ($/M tokens): 4.81 - Speed (tok/s): 89 - Latency (s): 104 - Context window: 400000 ### Kimi K2.5 - URL: https://subagent.developersdigest.tech/models/kimi-k2-5 - Creator: Kimi - Intelligence score: 47 - Price ($/M tokens): 1.2 - Speed (tok/s): 46 - Latency (s): 2.68 - Context window: 256000 ### GPT-5.2 (medium) - URL: https://subagent.developersdigest.tech/models/gpt-5-2-medium - Creator: OpenAI - Intelligence score: 47 - Price ($/M tokens): 4.81 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 400000 ### Claude Opus 4.6 - URL: https://subagent.developersdigest.tech/models/claude-opus-4-6 - Creator: Anthropic - Intelligence score: 46 - Price ($/M tokens): 10 - Speed (tok/s): 41 - Latency (s): 1.76 - Context window: 200000 ### Gemini 3 Flash - URL: https://subagent.developersdigest.tech/models/gemini-3-flash-reasoning - Creator: Google - Intelligence score: 46 - Price ($/M tokens): 1.13 - Speed (tok/s): 143 - Latency (s): 7.39 - Context window: 1000000 ### Qwen3.5 397B A17B - URL: https://subagent.developersdigest.tech/models/qwen3-5-397b-a17b - Creator: Alibaba - Intelligence score: 45 - Price ($/M tokens): 1.35 - Speed (tok/s): 100 - Latency (s): 2.67 - Context window: 262000 ### Claude Sonnet 4.6 - URL: https://subagent.developersdigest.tech/models/claude-sonnet-4-6 - Creator: Anthropic - Intelligence score: 44 - Price ($/M tokens): 6 - Speed (tok/s): 42 - Latency (s): 1.07 - Context window: 200000 ### Claude Sonnet 4.6 (Non-reasoning, Low Effort) - URL: https://subagent.developersdigest.tech/models/claude-sonnet-4-6-non-reasoning-low-effort - Creator: Anthropic - Intelligence score: 43 - Price ($/M tokens): 6 - Speed (tok/s): 43 - Latency (s): 1.03 - Context window: 200000 ### Qwen3.5 27B - URL: https://subagent.developersdigest.tech/models/qwen3-5-27b - Creator: Alibaba - Intelligence score: 42 - Price ($/M tokens): 0.82 - Speed (tok/s): 86 - Latency (s): 5.53 - Context window: 262000 ### MiniMax-M2.5 - URL: https://subagent.developersdigest.tech/models/minimax-m2-5 - Creator: MiniMax - Intelligence score: 42 - Price ($/M tokens): 0.53 - Speed (tok/s): 46 - Latency (s): 2.82 - Context window: 205000 ### DeepSeek V3.2 - URL: https://subagent.developersdigest.tech/models/deepseek-v3-2-reasoning - Creator: DeepSeek - Intelligence score: 42 - Price ($/M tokens): 0.32 - Speed (tok/s): 29 - Latency (s): 1.84 - Context window: 128000 ### Qwen3.5 122B A10B - URL: https://subagent.developersdigest.tech/models/qwen3-5-122b-a10b - Creator: Alibaba - Intelligence score: 42 - Price ($/M tokens): 1.1 - Speed (tok/s): 145 - Latency (s): 2.34 - Context window: 262000 ### Grok 4 - URL: https://subagent.developersdigest.tech/models/grok-4 - Creator: xAI - Intelligence score: 42 - Price ($/M tokens): 6 - Speed (tok/s): 39 - Latency (s): 10.4 - Context window: 256000 ### MiMo-V2-Flash (Feb 2026) - URL: https://subagent.developersdigest.tech/models/mimo-v2-0206 - Creator: Xiaomi - Intelligence score: 41 - Price ($/M tokens): 0.15 - Speed (tok/s): 120 - Latency (s): 2.05 - Context window: 256000 ### Gemini 3 Pro Preview (low) - URL: https://subagent.developersdigest.tech/models/gemini-3-pro-low - Creator: Google - Intelligence score: 41 - Price ($/M tokens): 4.5 - Speed (tok/s): 113 - Latency (s): 4.12 - Context window: 1000000 ### GPT-5 mini (high) - URL: https://subagent.developersdigest.tech/models/gpt-5-mini - Creator: OpenAI - Intelligence score: 41 - Price ($/M tokens): 0.69 - Speed (tok/s): 80 - Latency (s): 91.41 - Context window: 400000 ### GLM-5 - URL: https://subagent.developersdigest.tech/models/glm-5-non-reasoning - Creator: Z AI - Intelligence score: 41 - Price ($/M tokens): 1.55 - Speed (tok/s): 41 - Latency (s): 1.49 - Context window: 200000 ### Qwen3.5 397B A17B - URL: https://subagent.developersdigest.tech/models/qwen3-5-397b-a17b-non-reasoning - Creator: Alibaba - Intelligence score: 40 - Price ($/M tokens): 1.35 - Speed (tok/s): 97 - Latency (s): 2.64 - Context window: 262000 ### Qwen3 Max Thinking - URL: https://subagent.developersdigest.tech/models/qwen3-max-thinking - Creator: Alibaba - Intelligence score: 40 - Price ($/M tokens): 2.4 - Speed (tok/s): 35 - Latency (s): 3.93 - Context window: 256000 ### GPT-5.1 Codex mini (high) - URL: https://subagent.developersdigest.tech/models/gpt-5-1-codex-mini - Creator: OpenAI - Intelligence score: 39 - Price ($/M tokens): 0.69 - Speed (tok/s): 173 - Latency (s): 10.1 - Context window: 400000 ### Grok 4.1 Fast - URL: https://subagent.developersdigest.tech/models/grok-4-1-fast-reasoning - Creator: xAI - Intelligence score: 39 - Price ($/M tokens): 0.28 - Speed (tok/s): 113 - Latency (s): 8.83 - Context window: 2000000 ### o3 - URL: https://subagent.developersdigest.tech/models/o3 - Creator: OpenAI - Intelligence score: 38 - Price ($/M tokens): 3.5 - Speed (tok/s): 64 - Latency (s): 9.71 - Context window: 200000 ### Step 3.5 Flash - URL: https://subagent.developersdigest.tech/models/step-3-5-flash - Creator: StepFun - Intelligence score: 38 - Price ($/M tokens): 0.15 - Speed (tok/s): 158 - Latency (s): 1.51 - Context window: 256000 ### Kimi K2.5 - URL: https://subagent.developersdigest.tech/models/kimi-k2-5-non-reasoning - Creator: Kimi - Intelligence score: 37 - Price ($/M tokens): 1.2 - Speed (tok/s): 45 - Latency (s): 2.21 - Context window: 256000 ### Qwen3.5 27B - URL: https://subagent.developersdigest.tech/models/qwen3-5-27b-non-reasoning - Creator: Alibaba - Intelligence score: 37 - Price ($/M tokens): 0.82 - Speed (tok/s): 87 - Latency (s): 5.59 - Context window: 262000 ### Qwen3.5 35B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-5-35b-a3b - Creator: Alibaba - Intelligence score: 37 - Price ($/M tokens): 0.69 - Speed (tok/s): 161 - Latency (s): 2.04 - Context window: 262000 ### Claude 4.5 Haiku - URL: https://subagent.developersdigest.tech/models/claude-4-5-haiku-reasoning - Creator: Anthropic - Intelligence score: 37 - Price ($/M tokens): 2 - Speed (tok/s): 99 - Latency (s): 8.12 - Context window: 200000 ### KAT-Coder-Pro V1 - URL: https://subagent.developersdigest.tech/models/kat-coder-pro-v1 - Creator: KwaiKAT - Intelligence score: 36 - Price ($/M tokens): 0.53 - Speed (tok/s): 56 - Latency (s): 1.92 - Context window: 256000 ### Qwen3.5 122B A10B - URL: https://subagent.developersdigest.tech/models/qwen3-5-122b-a10b-non-reasoning - Creator: Alibaba - Intelligence score: 36 - Price ($/M tokens): 1.1 - Speed (tok/s): 137 - Latency (s): 2.3 - Context window: 262000 ### Nova 2.0 Pro Preview (medium) - URL: https://subagent.developersdigest.tech/models/nova-2-0-pro-reasoning-medium - Creator: Amazon - Intelligence score: 36 - Price ($/M tokens): 3.44 - Speed (tok/s): 142 - Latency (s): 11.86 - Context window: 256000 ### Gemini 3 Flash - URL: https://subagent.developersdigest.tech/models/gemini-3-flash - Creator: Google - Intelligence score: 35 - Price ($/M tokens): 1.13 - Speed (tok/s): 144 - Latency (s): 1.1 - Context window: 1000000 ### Gemini 2.5 Pro - URL: https://subagent.developersdigest.tech/models/gemini-2-5-pro - Creator: Google - Intelligence score: 35 - Price ($/M tokens): 3.44 - Speed (tok/s): 124 - Latency (s): 27.03 - Context window: 1000000 ### DeepSeek V3.2 Speciale - URL: https://subagent.developersdigest.tech/models/deepseek-v3-2-speciale - Creator: DeepSeek - Intelligence score: 34 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### GPT-5.2 - URL: https://subagent.developersdigest.tech/models/gpt-5-2-non-reasoning - Creator: OpenAI - Intelligence score: 34 - Price ($/M tokens): 4.81 - Speed (tok/s): 64 - Latency (s): 0.79 - Context window: 400000 ### Gemini 3.1 Flash-Lite Preview - URL: https://subagent.developersdigest.tech/models/gemini-3-1-flash-lite-preview - Creator: Google - Intelligence score: 34 - Price ($/M tokens): 0.56 - Speed (tok/s): 286 - Latency (s): 6.47 - Context window: 1000000 ### Doubao Seed Code - URL: https://subagent.developersdigest.tech/models/doubao-seed-code - Creator: ByteDance Seed - Intelligence score: 34 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 256000 ### gpt-oss-120B (high) - URL: https://subagent.developersdigest.tech/models/gpt-oss-120b - Creator: OpenAI - Intelligence score: 33 - Price ($/M tokens): 0.26 - Speed (tok/s): 262 - Latency (s): 0.91 - Context window: 131000 ### Mercury 2 - URL: https://subagent.developersdigest.tech/models/mercury-2 - Creator: Inception - Intelligence score: 33 - Price ($/M tokens): 0.38 - Speed (tok/s): 627 - Latency (s): 3.08 - Context window: 128000 ### Qwen3 Max Thinking (Preview) - URL: https://subagent.developersdigest.tech/models/qwen3-max-thinking-preview - Creator: Alibaba - Intelligence score: 32 - Price ($/M tokens): 2.4 - Speed (tok/s): 42 - Latency (s): 4.14 - Context window: 262000 ### Qwen3.5 9B - URL: https://subagent.developersdigest.tech/models/qwen3-5-9b - Creator: Alibaba - Intelligence score: 32 - Price ($/M tokens): 0.11 - Speed (tok/s): 123 - Latency (s): 0.72 - Context window: 262000 ### K-EXAONE - URL: https://subagent.developersdigest.tech/models/k-exaone - Creator: LG AI Research - Intelligence score: 32 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 256000 ### DeepSeek V3.2 - URL: https://subagent.developersdigest.tech/models/deepseek-v3-2 - Creator: DeepSeek - Intelligence score: 32 - Price ($/M tokens): 0.32 - Speed (tok/s): 29 - Latency (s): 1.65 - Context window: 128000 ### Grok 3 mini Reasoning (high) - URL: https://subagent.developersdigest.tech/models/grok-3-mini-reasoning - Creator: xAI - Intelligence score: 32 - Price ($/M tokens): 0.35 - Speed (tok/s): 191 - Latency (s): 0.63 - Context window: 1000000 ### Nova 2.0 Pro Preview (low) - URL: https://subagent.developersdigest.tech/models/nova-2-0-pro-reasoning-low - Creator: Amazon - Intelligence score: 32 - Price ($/M tokens): 3.44 - Speed (tok/s): 142 - Latency (s): 11.23 - Context window: 256000 ### Qwen3 Max - URL: https://subagent.developersdigest.tech/models/qwen3-max - Creator: Alibaba - Intelligence score: 31 - Price ($/M tokens): 2.4 - Speed (tok/s): 31 - Latency (s): 4.39 - Context window: 262000 ### Claude 4.5 Haiku - URL: https://subagent.developersdigest.tech/models/claude-4-5-haiku - Creator: Anthropic - Intelligence score: 31 - Price ($/M tokens): 2 - Speed (tok/s): 88 - Latency (s): 0.74 - Context window: 200000 ### Qwen3.5 35B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-5-35b-a3b-non-reasoning - Creator: Alibaba - Intelligence score: 31 - Price ($/M tokens): 0.69 - Speed (tok/s): 150 - Latency (s): 2.11 - Context window: 262000 ### MiMo-V2-Flash - URL: https://subagent.developersdigest.tech/models/mimo-v2-flash - Creator: Xiaomi - Intelligence score: 30 - Price ($/M tokens): 0.15 - Speed (tok/s): 120 - Latency (s): 1.96 - Context window: 256000 ### GLM-4.7-Flash - URL: https://subagent.developersdigest.tech/models/glm-4-7-flash - Creator: Z AI - Intelligence score: 30 - Price ($/M tokens): 0.15 - Speed (tok/s): 117 - Latency (s): 1 - Context window: 200000 ### Nova 2.0 Lite (medium) - URL: https://subagent.developersdigest.tech/models/nova-2-0-lite-reasoning-medium - Creator: Amazon - Intelligence score: 30 - Price ($/M tokens): 0.85 - Speed (tok/s): 177 - Latency (s): 15.47 - Context window: 1000000 ### Qwen3 235B A22B 2507 - URL: https://subagent.developersdigest.tech/models/qwen3-235b-a22b-instruct-2507-reasoning - Creator: Alibaba - Intelligence score: 30 - Price ($/M tokens): 2.63 - Speed (tok/s): 38 - Latency (s): 3.56 - Context window: 256000 ### ERNIE 5.0 Thinking Preview - URL: https://subagent.developersdigest.tech/models/ernie-5-0-thinking-preview - Creator: Baidu - Intelligence score: 29 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Grok Code Fast 1 - URL: https://subagent.developersdigest.tech/models/grok-code-fast-1 - Creator: xAI - Intelligence score: 29 - Price ($/M tokens): 0.53 - Speed (tok/s): 163 - Latency (s): 3.03 - Context window: 256000 ### Qwen3 Coder Next - URL: https://subagent.developersdigest.tech/models/qwen3-coder-next - Creator: Alibaba - Intelligence score: 28 - Price ($/M tokens): 0.53 - Speed (tok/s): 148 - Latency (s): 1.09 - Context window: 256000 ### Nova 2.0 Omni (medium) - URL: https://subagent.developersdigest.tech/models/nova-2-0-omni-reasoning-medium - Creator: Amazon - Intelligence score: 28 - Price ($/M tokens): 0.85 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 1000000 ### Qwen3 VL 235B A22B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-235b-a22b-reasoning - Creator: Alibaba - Intelligence score: 28 - Price ($/M tokens): 2.63 - Speed (tok/s): 41 - Latency (s): 3.14 - Context window: 262000 ### Apriel-v1.6-15B-Thinker - URL: https://subagent.developersdigest.tech/models/apriel-v1-6-15b-thinker - Creator: ServiceNow - Intelligence score: 28 - Price ($/M tokens): 0 - Speed (tok/s): 70 - Latency (s): 0.51 - Context window: 128000 ### Qwen3.5 9B - URL: https://subagent.developersdigest.tech/models/qwen3-5-9b-non-reasoning - Creator: Alibaba - Intelligence score: 27 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Magistral Medium 1.2 - URL: https://subagent.developersdigest.tech/models/magistral-medium-2509 - Creator: Mistral - Intelligence score: 27 - Price ($/M tokens): 2.75 - Speed (tok/s): 49 - Latency (s): 1.71 - Context window: 128000 ### Qwen3.5 4B - URL: https://subagent.developersdigest.tech/models/qwen3-5-4b - Creator: Alibaba - Intelligence score: 27 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### DeepSeek R1 0528 - URL: https://subagent.developersdigest.tech/models/deepseek-r1 - Creator: DeepSeek - Intelligence score: 27 - Price ($/M tokens): 2.36 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### GPT-5 nano (high) - URL: https://subagent.developersdigest.tech/models/gpt-5-nano - Creator: OpenAI - Intelligence score: 27 - Price ($/M tokens): 0.14 - Speed (tok/s): 127 - Latency (s): 107.5 - Context window: 400000 ### Qwen3 Next 80B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-next-80b-a3b-reasoning - Creator: Alibaba - Intelligence score: 27 - Price ($/M tokens): 1.88 - Speed (tok/s): 155 - Latency (s): 2.12 - Context window: 262000 ### Qwen3 235B 2507 - URL: https://subagent.developersdigest.tech/models/qwen3-235b-a22b-instruct-2507 - Creator: Alibaba - Intelligence score: 25 - Price ($/M tokens): 1.23 - Speed (tok/s): 53 - Latency (s): 2.39 - Context window: 256000 ### Qwen3 Coder 480B - URL: https://subagent.developersdigest.tech/models/qwen3-coder-480b-a35b-instruct - Creator: Alibaba - Intelligence score: 25 - Price ($/M tokens): 3 - Speed (tok/s): 58 - Latency (s): 3.12 - Context window: 262000 ### Qwen3 VL 32B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-32b-reasoning - Creator: Alibaba - Intelligence score: 25 - Price ($/M tokens): 2.63 - Speed (tok/s): 89 - Latency (s): 2.63 - Context window: 256000 ### Nova 2.0 Lite (low) - URL: https://subagent.developersdigest.tech/models/nova-2-0-lite-reasoning-low - Creator: Amazon - Intelligence score: 25 - Price ($/M tokens): 0.85 - Speed (tok/s): 177 - Latency (s): 7.39 - Context window: 1000000 ### DeepSeek V3 - URL: https://subagent.developersdigest.tech/models/deepseek-v3 - Creator: DeepSeek - Intelligence score: 25 - Price ($/M tokens): 0.27 - Speed (tok/s): 50 - Latency (s): 2.1 - Context window: 128000 ### gpt-oss-20B (high) - URL: https://subagent.developersdigest.tech/models/gpt-oss-20b - Creator: OpenAI - Intelligence score: 24 - Price ($/M tokens): 0.09 - Speed (tok/s): 285 - Latency (s): 0.71 - Context window: 131000 ### gpt-oss-120B (low) - URL: https://subagent.developersdigest.tech/models/gpt-oss-120b-low - Creator: OpenAI - Intelligence score: 24 - Price ($/M tokens): 0.26 - Speed (tok/s): 272 - Latency (s): 0.9 - Context window: 131000 ### NVIDIA Nemotron 3 Nano - URL: https://subagent.developersdigest.tech/models/nvidia-nemotron-3-nano-30b-a3b-reasoning - Creator: NVIDIA - Intelligence score: 24 - Price ($/M tokens): 0.1 - Speed (tok/s): 198 - Latency (s): 1 - Context window: 1000000 ### K2 Think V2 - URL: https://subagent.developersdigest.tech/models/k2-think-v2 - Creator: MBZUAI Institute of Foundation Models - Intelligence score: 24 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### HyperCLOVA X SEED Think (32B) - URL: https://subagent.developersdigest.tech/models/hyperclova-x-seed-think-32b - Creator: Naver - Intelligence score: 24 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Grok 4.1 Fast - URL: https://subagent.developersdigest.tech/models/grok-4-1-fast - Creator: xAI - Intelligence score: 24 - Price ($/M tokens): 0.28 - Speed (tok/s): 103 - Latency (s): 0.59 - Context window: 2000000 ### GLM-4.6V - URL: https://subagent.developersdigest.tech/models/glm-4-6v-reasoning - Creator: Z AI - Intelligence score: 23 - Price ($/M tokens): 0.45 - Speed (tok/s): 27 - Latency (s): 1.92 - Context window: 128000 ### K-EXAONE - URL: https://subagent.developersdigest.tech/models/k-exaone-non-reasoning - Creator: LG AI Research - Intelligence score: 23 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 256000 ### Nova 2.0 Omni (low) - URL: https://subagent.developersdigest.tech/models/nova-2-0-omni-reasoning-low - Creator: Amazon - Intelligence score: 23 - Price ($/M tokens): 0.85 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 1000000 ### Nova 2.0 Pro Preview - URL: https://subagent.developersdigest.tech/models/nova-2-0-pro - Creator: Amazon - Intelligence score: 23 - Price ($/M tokens): 3.44 - Speed (tok/s): 128 - Latency (s): 0.79 - Context window: 256000 ### Mi:dm K 2.5 Pro - URL: https://subagent.developersdigest.tech/models/mi-dm-k-2-5-pro-dec28 - Creator: Korea Telecom - Intelligence score: 23 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Mistral Large 3 - URL: https://subagent.developersdigest.tech/models/mistral-large-3 - Creator: Mistral - Intelligence score: 23 - Price ($/M tokens): 0.75 - Speed (tok/s): 48 - Latency (s): 0.99 - Context window: 256000 ### Ring-1T - URL: https://subagent.developersdigest.tech/models/ring-1t - Creator: InclusionAI - Intelligence score: 23 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### DeepSeek R1 (Jan 2025) - URL: https://subagent.developersdigest.tech/models/deepseek-r1-0120 - Creator: DeepSeek - Intelligence score: 23 - Price ($/M tokens): 2.19 - Speed (tok/s): 28 - Latency (s): 3.5 - Context window: 128000 ### Qwen3.5 4B - URL: https://subagent.developersdigest.tech/models/qwen3-5-4b-non-reasoning - Creator: Alibaba - Intelligence score: 23 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Qwen3 30B A3B 2507 - URL: https://subagent.developersdigest.tech/models/qwen3-30b-a3b-2507-reasoning - Creator: Alibaba - Intelligence score: 22 - Price ($/M tokens): 0.75 - Speed (tok/s): 123 - Latency (s): 2.44 - Context window: 262000 ### INTELLECT-3 - URL: https://subagent.developersdigest.tech/models/intellect-3 - Creator: Prime Intellect - Intelligence score: 22 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 131000 ### GLM-4.7-Flash - URL: https://subagent.developersdigest.tech/models/glm-4-7-flash-non-reasoning - Creator: Z AI - Intelligence score: 22 - Price ($/M tokens): 0.15 - Speed (tok/s): 168 - Latency (s): 1.26 - Context window: 200000 ### Devstral 2 - URL: https://subagent.developersdigest.tech/models/devstral-2 - Creator: Mistral - Intelligence score: 22 - Price ($/M tokens): 0 - Speed (tok/s): 78 - Latency (s): 0.82 - Context window: 256000 ### Solar Open 100B - URL: https://subagent.developersdigest.tech/models/solar-open-100b-reasoning - Creator: Upstage - Intelligence score: 22 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Gemini 2.5 Flash-Lite (Sep) - URL: https://subagent.developersdigest.tech/models/gemini-2-5-flash-lite-preview-09-2025-reasoning - Creator: Google - Intelligence score: 22 - Price ($/M tokens): 0.17 - Speed (tok/s): 337 - Latency (s): 17.84 - Context window: 1000000 ### Mistral Medium 3.1 - URL: https://subagent.developersdigest.tech/models/mistral-medium-3-1 - Creator: Mistral - Intelligence score: 21 - Price ($/M tokens): 0.8 - Speed (tok/s): 98 - Latency (s): 1.18 - Context window: 128000 ### gpt-oss-20B (low) - URL: https://subagent.developersdigest.tech/models/gpt-oss-20b-low - Creator: OpenAI - Intelligence score: 21 - Price ($/M tokens): 0.09 - Speed (tok/s): 292 - Latency (s): 0.7 - Context window: 131000 ### Qwen3 VL 235B A22B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-235b-a22b-instruct - Creator: Alibaba - Intelligence score: 21 - Price ($/M tokens): 1.23 - Speed (tok/s): 56 - Latency (s): 2.66 - Context window: 262000 ### K2-V2 (high) - URL: https://subagent.developersdigest.tech/models/k2-v2 - Creator: MBZUAI Institute of Foundation Models - Intelligence score: 21 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 512000 ### Qwen3 Next 80B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-next-80b-a3b-instruct - Creator: Alibaba - Intelligence score: 20 - Price ($/M tokens): 0.88 - Speed (tok/s): 157 - Latency (s): 2.09 - Context window: 262000 ### Tri-21B-think Preview - URL: https://subagent.developersdigest.tech/models/tri-21b-think-preview - Creator: Trillion Labs - Intelligence score: 20 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Qwen3 Coder 30B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-coder-30b-a3b-instruct - Creator: Alibaba - Intelligence score: 20 - Price ($/M tokens): 0.9 - Speed (tok/s): 26 - Latency (s): 2.7 - Context window: 262000 ### Qwen3 VL 30B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-30b-a3b-reasoning - Creator: Alibaba - Intelligence score: 20 - Price ($/M tokens): 0.75 - Speed (tok/s): 90 - Latency (s): 2.29 - Context window: 256000 ### Devstral Small 2 - URL: https://subagent.developersdigest.tech/models/devstral-small-2 - Creator: Mistral - Intelligence score: 19 - Price ($/M tokens): 0 - Speed (tok/s): 197 - Latency (s): 0.55 - Context window: 256000 ### Gemini 2.5 Flash-Lite (Sep) - URL: https://subagent.developersdigest.tech/models/gemini-2-5-flash-lite-preview-09-2025 - Creator: Google - Intelligence score: 19 - Price ($/M tokens): 0.17 - Speed (tok/s): 288 - Latency (s): 3.48 - Context window: 1000000 ### Motif-2-12.7B - URL: https://subagent.developersdigest.tech/models/motif-2-12-7b - Creator: Motif Technologies - Intelligence score: 19 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Ling-1T - URL: https://subagent.developersdigest.tech/models/ling-1t - Creator: InclusionAI - Intelligence score: 19 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Nova Premier - URL: https://subagent.developersdigest.tech/models/nova-premier - Creator: Amazon - Intelligence score: 19 - Price ($/M tokens): 5 - Speed (tok/s): 25 - Latency (s): 2.83 - Context window: 1000000 ### NVIDIA Nemotron Nano 9B V2 - URL: https://subagent.developersdigest.tech/models/nvidia-nemotron-nano-9b-v2 - Creator: NVIDIA - Intelligence score: 19 - Price ($/M tokens): 0.1 - Speed (tok/s): 130 - Latency (s): 1.55 - Context window: 131000 ### Llama Nemotron Super 49B v1.5 - URL: https://subagent.developersdigest.tech/models/llama-nemotron-super-49b-v1-5-reasoning - Creator: NVIDIA - Intelligence score: 19 - Price ($/M tokens): 0.17 - Speed (tok/s): 81 - Latency (s): 0.81 - Context window: 128000 ### K2-V2 (medium) - URL: https://subagent.developersdigest.tech/models/k2-v2-medium - Creator: MBZUAI Institute of Foundation Models - Intelligence score: 19 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 512000 ### Tri-21B-Think - URL: https://subagent.developersdigest.tech/models/tri-21b-think-v0-5 - Creator: Trillion Labs - Intelligence score: 19 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Hermes 4 405B - URL: https://subagent.developersdigest.tech/models/hermes-4-llama-3-1-405b-reasoning - Creator: Nous Research - Intelligence score: 19 - Price ($/M tokens): 1.5 - Speed (tok/s): 33 - Latency (s): 2.4 - Context window: 128000 ### Llama 3.3 Nemotron Super 49B - URL: https://subagent.developersdigest.tech/models/llama-3-3-nemotron-super-49b-reasoning - Creator: NVIDIA - Intelligence score: 18 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Llama 4 Maverick - URL: https://subagent.developersdigest.tech/models/llama-4-maverick - Creator: Meta - Intelligence score: 18 - Price ($/M tokens): 0.46 - Speed (tok/s): 123 - Latency (s): 0.79 - Context window: 1000000 ### Qwen3 4B 2507 - URL: https://subagent.developersdigest.tech/models/qwen3-4b-2507-instruct-reasoning - Creator: Alibaba - Intelligence score: 18 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Magistral Small 1.2 - URL: https://subagent.developersdigest.tech/models/magistral-small-2509 - Creator: Mistral - Intelligence score: 18 - Price ($/M tokens): 0.75 - Speed (tok/s): 128 - Latency (s): 0.77 - Context window: 128000 ### Nova 2.0 Lite - URL: https://subagent.developersdigest.tech/models/nova-2-0-lite - Creator: Amazon - Intelligence score: 18 - Price ($/M tokens): 0.85 - Speed (tok/s): 179 - Latency (s): 1.07 - Context window: 1000000 ### Hermes 4 405B - URL: https://subagent.developersdigest.tech/models/hermes-4-llama-3-1-405b - Creator: Nous Research - Intelligence score: 18 - Price ($/M tokens): 1.5 - Speed (tok/s): 33 - Latency (s): 2.44 - Context window: 128000 ### Llama 3.1 405B - URL: https://subagent.developersdigest.tech/models/llama-3-1-instruct-405b - Creator: Meta - Intelligence score: 17 - Price ($/M tokens): 4.38 - Speed (tok/s): 25 - Latency (s): 2.73 - Context window: 128000 ### Qwen3 VL 32B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-32b-instruct - Creator: Alibaba - Intelligence score: 17 - Price ($/M tokens): 1.23 - Speed (tok/s): 72 - Latency (s): 2.63 - Context window: 256000 ### GLM-4.6V - URL: https://subagent.developersdigest.tech/models/glm-4-6v - Creator: Z AI - Intelligence score: 17 - Price ($/M tokens): 0.45 - Speed (tok/s): 20 - Latency (s): 11.49 - Context window: 128000 ### EXAONE 4.0 32B - URL: https://subagent.developersdigest.tech/models/exaone-4-0-32b-reasoning - Creator: LG AI Research - Intelligence score: 17 - Price ($/M tokens): 0.7 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 131000 ### Qwen3 VL 8B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-8b-reasoning - Creator: Alibaba - Intelligence score: 17 - Price ($/M tokens): 0.66 - Speed (tok/s): 114 - Latency (s): 2.32 - Context window: 256000 ### Nova 2.0 Omni - URL: https://subagent.developersdigest.tech/models/nova-2-0-omni - Creator: Amazon - Intelligence score: 17 - Price ($/M tokens): 0.85 - Speed (tok/s): 169 - Latency (s): 1.15 - Context window: 1000000 ### DeepSeek R1 0528 Qwen3 8B - URL: https://subagent.developersdigest.tech/models/deepseek-r1-qwen3-8b - Creator: DeepSeek - Intelligence score: 16 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 33000 ### Qwen3.5 2B - URL: https://subagent.developersdigest.tech/models/qwen3-5-2b - Creator: Alibaba - Intelligence score: 16 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Qwen3 VL 30B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-30b-a3b-instruct - Creator: Alibaba - Intelligence score: 16 - Price ($/M tokens): 0.35 - Speed (tok/s): 108 - Latency (s): 2.18 - Context window: 256000 ### Hermes 4 70B - URL: https://subagent.developersdigest.tech/models/hermes-4-llama-3-1-70b-reasoning - Creator: Nous Research - Intelligence score: 16 - Price ($/M tokens): 0.2 - Speed (tok/s): 79 - Latency (s): 1.35 - Context window: 128000 ### Ministral 3 14B - URL: https://subagent.developersdigest.tech/models/ministral-3-14b - Creator: Mistral - Intelligence score: 16 - Price ($/M tokens): 0.2 - Speed (tok/s): 133 - Latency (s): 0.61 - Context window: 256000 ### DeepSeek R1 Distill Llama 70B - URL: https://subagent.developersdigest.tech/models/deepseek-r1-distill-llama-70b - Creator: DeepSeek - Intelligence score: 16 - Price ($/M tokens): 0.88 - Speed (tok/s): 59 - Latency (s): 2.76 - Context window: 128000 ### Falcon-H1R-7B - URL: https://subagent.developersdigest.tech/models/falcon-h1r-7b - Creator: TII UAE - Intelligence score: 16 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 256000 ### Ling-flash-2.0 - URL: https://subagent.developersdigest.tech/models/ling-flash-2-0 - Creator: InclusionAI - Intelligence score: 16 - Price ($/M tokens): 0.25 - Speed (tok/s): 65 - Latency (s): 2.25 - Context window: 128000 ### Qwen3 Omni 30B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-omni-30b-a3b-reasoning - Creator: Alibaba - Intelligence score: 16 - Price ($/M tokens): 0.43 - Speed (tok/s): 91 - Latency (s): 1.92 - Context window: 66000 ### Step3 VL 10B - URL: https://subagent.developersdigest.tech/models/step3-vl-10b - Creator: StepFun - Intelligence score: 15 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 66000 ### Mistral Small 3.2 - URL: https://subagent.developersdigest.tech/models/mistral-small-3-2 - Creator: Mistral - Intelligence score: 15 - Price ($/M tokens): 0.15 - Speed (tok/s): 150 - Latency (s): 0.56 - Context window: 128000 ### Llama Nemotron Ultra - URL: https://subagent.developersdigest.tech/models/llama-3-1-nemotron-ultra-253b-v1-reasoning - Creator: NVIDIA - Intelligence score: 15 - Price ($/M tokens): 0.9 - Speed (tok/s): 42 - Latency (s): 2.52 - Context window: 128000 ### Qwen3 30B A3B 2507 - URL: https://subagent.developersdigest.tech/models/qwen3-30b-a3b-2507 - Creator: Alibaba - Intelligence score: 15 - Price ($/M tokens): 0.35 - Speed (tok/s): 70 - Latency (s): 1.9 - Context window: 262000 ### ERNIE 4.5 300B A47B - URL: https://subagent.developersdigest.tech/models/ernie-4-5-300b-a47b - Creator: Baidu - Intelligence score: 15 - Price ($/M tokens): 0.48 - Speed (tok/s): 24 - Latency (s): 3.65 - Context window: 131000 ### Solar Pro 2 - URL: https://subagent.developersdigest.tech/models/solar-pro-2-reasoning - Creator: Upstage - Intelligence score: 15 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 66000 ### NVIDIA Nemotron Nano 12B v2 VL - URL: https://subagent.developersdigest.tech/models/nvidia-nemotron-nano-12b-v2-vl-reasoning - Creator: NVIDIA - Intelligence score: 15 - Price ($/M tokens): 0.3 - Speed (tok/s): 131 - Latency (s): 1.05 - Context window: 128000 ### Ministral 3 8B - URL: https://subagent.developersdigest.tech/models/ministral-3-8b - Creator: Mistral - Intelligence score: 15 - Price ($/M tokens): 0.15 - Speed (tok/s): 161 - Latency (s): 0.51 - Context window: 256000 ### NVIDIA Nemotron Nano 9B V2 - URL: https://subagent.developersdigest.tech/models/nvidia-nemotron-nano-9b-v2-reasoning - Creator: NVIDIA - Intelligence score: 15 - Price ($/M tokens): 0.07 - Speed (tok/s): 109 - Latency (s): 1.1 - Context window: 131000 ### Qwen3.5 2B - URL: https://subagent.developersdigest.tech/models/qwen3-5-2b-non-reasoning - Creator: Alibaba - Intelligence score: 15 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Llama Nemotron Super 49B v1.5 - URL: https://subagent.developersdigest.tech/models/llama-nemotron-super-49b-v1-5 - Creator: NVIDIA - Intelligence score: 15 - Price ($/M tokens): 0.17 - Speed (tok/s): 81 - Latency (s): 0.61 - Context window: 128000 ### Llama 3.3 70B - URL: https://subagent.developersdigest.tech/models/llama-3-3-instruct-70b - Creator: Meta - Intelligence score: 14 - Price ($/M tokens): 0.68 - Speed (tok/s): 87 - Latency (s): 1.4 - Context window: 128000 ### K2-V2 (low) - URL: https://subagent.developersdigest.tech/models/k2-v2-low - Creator: MBZUAI Institute of Foundation Models - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 512000 ### Llama 3.1 Nemotron Nano 4B v1.1 - URL: https://subagent.developersdigest.tech/models/llama-3-1-nemotron-nano-4b-reasoning - Creator: NVIDIA - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Kimi Linear 48B A3B Instruct - URL: https://subagent.developersdigest.tech/models/kimi-linear-48b-a3b-instruct - Creator: Kimi - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 1000000 ### Llama 3.3 Nemotron Super 49B - URL: https://subagent.developersdigest.tech/models/llama-3-3-nemotron-super-49b - Creator: NVIDIA - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Qwen3 VL 8B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-8b-instruct - Creator: Alibaba - Intelligence score: 14 - Price ($/M tokens): 0.31 - Speed (tok/s): 112 - Latency (s): 2.31 - Context window: 256000 ### Ring-flash-2.0 - URL: https://subagent.developersdigest.tech/models/ring-flash-2-0 - Creator: InclusionAI - Intelligence score: 14 - Price ($/M tokens): 0.25 - Speed (tok/s): 82 - Latency (s): 2.54 - Context window: 128000 ### Olmo 3.1 32B Think - URL: https://subagent.developersdigest.tech/models/olmo-3-1-32b-think - Creator: Allen Institute for AI - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 85 - Latency (s): 1.75 - Context window: 66000 ### Qwen3 VL 4B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-4b-reasoning - Creator: Alibaba - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 256000 ### Solar Pro 2 - URL: https://subagent.developersdigest.tech/models/solar-pro-2 - Creator: Upstage - Intelligence score: 14 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 66000 ### Llama 4 Scout - URL: https://subagent.developersdigest.tech/models/llama-4-scout - Creator: Meta - Intelligence score: 14 - Price ($/M tokens): 0.29 - Speed (tok/s): 128 - Latency (s): 0.77 - Context window: 10000000 ### Command A - URL: https://subagent.developersdigest.tech/models/command-a - Creator: Cohere - Intelligence score: 13 - Price ($/M tokens): 4.38 - Speed (tok/s): 44 - Latency (s): 1.81 - Context window: 256000 ### Llama 3.1 Nemotron 70B - URL: https://subagent.developersdigest.tech/models/llama-3-1-nemotron-instruct-70b - Creator: NVIDIA - Intelligence score: 13 - Price ($/M tokens): 1.2 - Speed (tok/s): 34 - Latency (s): 2.15 - Context window: 128000 ### NVIDIA Nemotron 3 Nano - URL: https://subagent.developersdigest.tech/models/nvidia-nemotron-3-nano-30b-a3b - Creator: NVIDIA - Intelligence score: 13 - Price ($/M tokens): 0.09 - Speed (tok/s): 183 - Latency (s): 0.97 - Context window: 1000000 ### Qwen3 4B 2507 - URL: https://subagent.developersdigest.tech/models/qwen3-4b-2507-instruct - Creator: Alibaba - Intelligence score: 13 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Hermes 4 70B - URL: https://subagent.developersdigest.tech/models/hermes-4-llama-3-1-70b - Creator: Nous Research - Intelligence score: 13 - Price ($/M tokens): 0.2 - Speed (tok/s): 80 - Latency (s): 1.35 - Context window: 128000 ### Olmo 3.1 32B Instruct - URL: https://subagent.developersdigest.tech/models/olmo-3-1-32b-instruct - Creator: Allen Institute for AI - Intelligence score: 12 - Price ($/M tokens): 0.3 - Speed (tok/s): 52 - Latency (s): 0.9 - Context window: 66000 ### R1 1776 - URL: https://subagent.developersdigest.tech/models/r1-1776 - Creator: Perplexity - Intelligence score: 12 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Llama 3.2 90B (Vision) - URL: https://subagent.developersdigest.tech/models/llama-3-2-instruct-90b-vision - Creator: Meta - Intelligence score: 12 - Price ($/M tokens): 0.72 - Speed (tok/s): 60 - Latency (s): 0.92 - Context window: 128000 ### EXAONE 4.0 32B - URL: https://subagent.developersdigest.tech/models/exaone-4-0-32b - Creator: LG AI Research - Intelligence score: 12 - Price ($/M tokens): 0.7 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 131000 ### Ministral 3 3B - URL: https://subagent.developersdigest.tech/models/ministral-3-3b - Creator: Mistral - Intelligence score: 11 - Price ($/M tokens): 0.1 - Speed (tok/s): 252 - Latency (s): 0.42 - Context window: 256000 ### Phi-4 Mini - URL: https://subagent.developersdigest.tech/models/phi-4-mini - Creator: Microsoft Azure - Intelligence score: 11 - Price ($/M tokens): 0 - Speed (tok/s): 41 - Latency (s): 0.81 - Context window: 128000 ### DeepHermes 3 - Mistral 24B - URL: https://subagent.developersdigest.tech/models/deephermes-3-mistral-24b-preview - Creator: Nous Research - Intelligence score: 11 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Jamba 1.7 Large - URL: https://subagent.developersdigest.tech/models/jamba-1-7-large - Creator: AI21 Labs - Intelligence score: 11 - Price ($/M tokens): 3.5 - Speed (tok/s): 61 - Latency (s): 1.26 - Context window: 256000 ### Granite 4.0 H Small - URL: https://subagent.developersdigest.tech/models/granite-4-0-h-small - Creator: IBM - Intelligence score: 11 - Price ($/M tokens): 0.11 - Speed (tok/s): 372 - Latency (s): 10.24 - Context window: 128000 ### Qwen3 Omni 30B A3B - URL: https://subagent.developersdigest.tech/models/qwen3-omni-30b-a3b-instruct - Creator: Alibaba - Intelligence score: 11 - Price ($/M tokens): 0.43 - Speed (tok/s): 91 - Latency (s): 1.82 - Context window: 66000 ### LFM2 24B A2B - URL: https://subagent.developersdigest.tech/models/lfm2-24b-a2b - Creator: Liquid AI - Intelligence score: 10 - Price ($/M tokens): 0.05 - Speed (tok/s): 241 - Latency (s): 0.48 - Context window: 33000 ### Phi-4 - URL: https://subagent.developersdigest.tech/models/phi-4 - Creator: Microsoft Azure - Intelligence score: 10 - Price ($/M tokens): 0.22 - Speed (tok/s): 8 - Latency (s): 3.51 - Context window: 16000 ### Gemma 3 27B - URL: https://subagent.developersdigest.tech/models/gemma-3-27b - Creator: Google - Intelligence score: 10 - Price ($/M tokens): 0 - Speed (tok/s): 35 - Latency (s): 2.03 - Context window: 128000 ### Nova Micro - URL: https://subagent.developersdigest.tech/models/nova-micro - Creator: Amazon - Intelligence score: 10 - Price ($/M tokens): 0.06 - Speed (tok/s): 280 - Latency (s): 0.66 - Context window: 130000 ### NVIDIA Nemotron Nano 12B v2 VL - URL: https://subagent.developersdigest.tech/models/nvidia-nemotron-nano-12b-v2-vl - Creator: NVIDIA - Intelligence score: 10 - Price ($/M tokens): 0.3 - Speed (tok/s): 135 - Latency (s): 1.13 - Context window: 128000 ### Phi-4 Multimodal - URL: https://subagent.developersdigest.tech/models/phi-4-multimodal - Creator: Microsoft Azure - Intelligence score: 10 - Price ($/M tokens): 0 - Speed (tok/s): 17 - Latency (s): 0.82 - Context window: 128000 ### Jamba Reasoning 3B - URL: https://subagent.developersdigest.tech/models/jamba-reasoning-3b - Creator: AI21 Labs - Intelligence score: 10 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Qwen3 VL 4B - URL: https://subagent.developersdigest.tech/models/qwen3-vl-4b-instruct - Creator: Alibaba - Intelligence score: 10 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 256000 ### Reka Flash 3 - URL: https://subagent.developersdigest.tech/models/reka-flash-3 - Creator: Reka AI - Intelligence score: 10 - Price ($/M tokens): 0.35 - Speed (tok/s): 42 - Latency (s): 2.39 - Context window: 128000 ### Olmo 3 7B Think - URL: https://subagent.developersdigest.tech/models/olmo-3-7b-think - Creator: Allen Institute for AI - Intelligence score: 9 - Price ($/M tokens): 0.14 - Speed (tok/s): 64 - Latency (s): 1.03 - Context window: 66000 ### Ling-mini-2.0 - URL: https://subagent.developersdigest.tech/models/ling-mini-2-0 - Creator: InclusionAI - Intelligence score: 9 - Price ($/M tokens): 0.12 - Speed (tok/s): 158 - Latency (s): 2.34 - Context window: 131000 ### Qwen3.5 0.8B - URL: https://subagent.developersdigest.tech/models/qwen3-5-0-8b - Creator: Alibaba - Intelligence score: 9 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 262000 ### Gemma 3 12B - URL: https://subagent.developersdigest.tech/models/gemma-3-12b - Creator: Google - Intelligence score: 9 - Price ($/M tokens): 0 - Speed (tok/s): 34 - Latency (s): 28.01 - Context window: 128000 ### Llama 3.2 11B (Vision) - URL: https://subagent.developersdigest.tech/models/llama-3-2-instruct-11b-vision - Creator: Meta - Intelligence score: 9 - Price ($/M tokens): 0.16 - Speed (tok/s): 44 - Latency (s): 0.6 - Context window: 128000 ### Exaone 4.0 1.2B - URL: https://subagent.developersdigest.tech/models/exaone-4-0-1-2b-reasoning - Creator: LG AI Research - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 64000 ### Olmo 3 7B - URL: https://subagent.developersdigest.tech/models/olmo-3-7b-instruct - Creator: Allen Institute for AI - Intelligence score: 8 - Price ($/M tokens): 0.13 - Speed (tok/s): 137 - Latency (s): 0.85 - Context window: 66000 ### Exaone 4.0 1.2B - URL: https://subagent.developersdigest.tech/models/exaone-4-0-1-2b - Creator: LG AI Research - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 64000 ### LFM2.5-1.2B-Thinking - URL: https://subagent.developersdigest.tech/models/lfm2-5-1-2b-thinking - Creator: Liquid AI - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Jamba 1.7 Mini - URL: https://subagent.developersdigest.tech/models/jamba-1-7-mini - Creator: AI21 Labs - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 258000 ### LFM2 2.6B - URL: https://subagent.developersdigest.tech/models/lfm2-2-6b - Creator: Liquid AI - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 33000 ### LFM2.5-1.2B-Instruct - URL: https://subagent.developersdigest.tech/models/lfm2-5-1-2b-instruct - Creator: Liquid AI - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Granite 4.0 H 1B - URL: https://subagent.developersdigest.tech/models/granite-4-0-h-nano-1b - Creator: IBM - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Qwen3 1.7B - URL: https://subagent.developersdigest.tech/models/qwen3-1-7b - Creator: Alibaba - Intelligence score: 8 - Price ($/M tokens): 0.4 - Speed (tok/s): 125 - Latency (s): 1.74 - Context window: 32000 ### Gemma 3 270M - URL: https://subagent.developersdigest.tech/models/gemma-3-270m - Creator: Google - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Granite 4.0 Micro - URL: https://subagent.developersdigest.tech/models/granite-4-0-micro - Creator: IBM - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### DeepHermes 3 - Llama-3.1 8B - URL: https://subagent.developersdigest.tech/models/deephermes-3-llama-3-1-8b-preview - Creator: Nous Research - Intelligence score: 8 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### Granite 4.0 1B - URL: https://subagent.developersdigest.tech/models/granite-4-0-nano-1b - Creator: IBM - Intelligence score: 7 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 128000 ### LFM2 8B A1B - URL: https://subagent.developersdigest.tech/models/lfm2-8b-a1b - Creator: Liquid AI - Intelligence score: 7 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 33000 ### Qwen3 1.7B - URL: https://subagent.developersdigest.tech/models/qwen3-1-7b - Creator: Alibaba - Intelligence score: 7 - Price ($/M tokens): 0.19 - Speed (tok/s): 126 - Latency (s): 1.71 - Context window: 32000 ### Qwen3 0.6B - URL: https://subagent.developersdigest.tech/models/qwen3-0-6b - Creator: Alibaba - Intelligence score: 6 - Price ($/M tokens): 0.4 - Speed (tok/s): 191 - Latency (s): 1.6 - Context window: 32000 ### Gemma 3n E4B - URL: https://subagent.developersdigest.tech/models/gemma-3n-e4b - Creator: Google - Intelligence score: 6 - Price ($/M tokens): 0.03 - Speed (tok/s): 51 - Latency (s): 1.11 - Context window: 32000 ### Gemma 3 4B - URL: https://subagent.developersdigest.tech/models/gemma-3-4b - Creator: Google - Intelligence score: 6 - Price ($/M tokens): 0 - Speed (tok/s): 38 - Latency (s): 1.45 - Context window: 128000 ### LFM2.5-VL-1.6B - URL: https://subagent.developersdigest.tech/models/lfm2-5-vl-1-6b - Creator: Liquid AI - Intelligence score: 6 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ### Granite 4.0 350M - URL: https://subagent.developersdigest.tech/models/granite-4-0-350m - Creator: IBM - Intelligence score: 6 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 33000 ### Qwen3 0.6B - URL: https://subagent.developersdigest.tech/models/qwen3-0-6b - Creator: Alibaba - Intelligence score: 6 - Price ($/M tokens): 0.19 - Speed (tok/s): 193 - Latency (s): 1.58 - Context window: 32000 ### Gemma 3 1B - URL: https://subagent.developersdigest.tech/models/gemma-3-1b - Creator: Google - Intelligence score: 6 - Price ($/M tokens): 0 - Speed (tok/s): 52 - Latency (s): 0.75 - Context window: 32000 ### Granite 4.0 H 350M - URL: https://subagent.developersdigest.tech/models/granite-4-0-h-350m - Creator: IBM - Intelligence score: 5 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 33000 ### Gemma 3n E2B - URL: https://subagent.developersdigest.tech/models/gemma-3n-e2b - Creator: Google - Intelligence score: 5 - Price ($/M tokens): 0 - Speed (tok/s): 0 - Latency (s): 0 - Context window: 32000 ## Frameworks (20) ### LangChain - URL: https://subagent.developersdigest.tech/frameworks#langchain - GitHub: https://github.com/langchain-ai/langchain - npm: langchain - pypi: langchain - Category: Full-Stack - Language: Both - Stars: 130k+ - Creator: LangChain - Best for: Teams needing a batteries-included framework with broad LLM provider support and production tracing. - Features: Composable chains and agents, Built-in memory systems, RAG and retrieval pipelines, Tool calling and function execution, Document loaders and text splitters, Multi-provider LLM support, Structured output parsing, LangSmith tracing integration Full-stack framework for building LLM-powered applications with composable chains, agents, memory, and retrieval. ### LangGraph - URL: https://subagent.developersdigest.tech/frameworks#langgraph - GitHub: https://github.com/langchain-ai/langgraph - npm: @langchain/langgraph - pypi: langgraph - Category: Multi-Agent - Language: Both - Stars: 25k+ - Creator: LangChain - Best for: Complex multi-agent workflows that need state management, branching logic, and production deployment. - Features: Stateful graph-based agent flows, Cyclic and branching execution, Built-in persistence and checkpointing, Human-in-the-loop breakpoints, Streaming support, Subgraph composition, LangGraph Platform for deployment, Pre-built ReAct agent template Library for building stateful, multi-agent applications as cyclic graphs with built-in persistence and human-in-the-loop support. ### CrewAI - URL: https://subagent.developersdigest.tech/frameworks#crewai - GitHub: https://github.com/crewAIInc/crewAI - pypi: crewai - Category: Multi-Agent - Language: Python - Stars: 45k+ - Creator: CrewAI - Best for: Teams wanting an intuitive role-based approach to multi-agent systems with minimal boilerplate. - Features: Role-based agent design, Sequential and parallel task execution, Built-in tool ecosystem, Agent delegation and collaboration, Memory and context sharing, Process types: sequential, hierarchical, CrewAI Enterprise for deployment, Kickoff with inputs for dynamic tasks Role-based multi-agent orchestration framework where AI agents collaborate as a crew with defined roles, goals, and backstories. ### OpenAI Agents SDK - URL: https://subagent.developersdigest.tech/frameworks#openai-agents-sdk - GitHub: https://github.com/openai/openai-agents-python - pypi: openai-agents - Category: SDK - Language: Python - Stars: 18k+ - Creator: OpenAI - Best for: OpenAI-centric teams wanting first-party agent primitives with handoffs and safety guardrails. - Features: Agent handoffs between specialists, Input and output guardrails, Built-in tracing and observability, Tool calling with type safety, Streaming responses, Context management, MCP server integration, Lifecycle hooks for customization Official OpenAI framework for building agentic applications with handoffs, guardrails, and tracing built in. ### Anthropic SDK - URL: https://subagent.developersdigest.tech/frameworks#anthropic-sdk - GitHub: https://github.com/anthropics/anthropic-sdk-python - npm: @anthropic-ai/sdk - pypi: anthropic - Category: SDK - Language: Both - Stars: 3k+ - Creator: Anthropic - Best for: Teams building with Claude who want direct SDK access to tool use, extended thinking, and computer use capabilities. - Features: Tool use and function calling, Extended thinking for complex reasoning, Multi-turn agent loops, Streaming with tool use, Computer use (beta), MCP client support, Prompt caching for efficiency, Batch API for high-volume tasks Official Anthropic SDK for building agents with Claude, featuring tool use, multi-turn conversations, and extended thinking. ### Vercel AI SDK - URL: https://subagent.developersdigest.tech/frameworks#vercel-ai-sdk - GitHub: https://github.com/vercel/ai - npm: ai - Category: Full-Stack - Language: TypeScript - Stars: 22k+ - Creator: Vercel - Best for: TypeScript/Next.js developers building full-stack AI apps with streaming UIs and multi-provider support. - Features: Streaming UI with React Server Components, Multi-step tool calling agents, Structured output with Zod schemas, Multi-provider support (OpenAI, Anthropic, Google, etc.), useChat and useCompletion React hooks, Generative UI rendering, Text and object generation, MCP client integration Full-stack TypeScript toolkit for building AI applications with streaming, tool calling, multi-step agents, and structured output. ### AutoGen - URL: https://subagent.developersdigest.tech/frameworks#autogen - GitHub: https://github.com/microsoft/autogen - pypi: autogen-agentchat - Category: Multi-Agent - Language: Python - Stars: 55k+ - Creator: Microsoft - Best for: Research teams and enterprises needing flexible multi-agent conversations with code execution capabilities. - Features: Multi-agent conversations, Code generation and execution, Human-in-the-loop integration, Flexible agent topologies, Group chat with multiple agents, Custom agent creation, Nested agent chats, AutoGen Studio visual builder Microsoft framework for building multi-agent conversational systems where agents can chat with each other to solve tasks. ### Composio - URL: https://subagent.developersdigest.tech/frameworks#composio - GitHub: https://github.com/ComposioHQ/composio - npm: composio-core - pypi: composio-core - Category: Tool Integration - Language: Both - Stars: 27k+ - Creator: Composio - Best for: Teams that need agents to interact with real-world SaaS tools without building individual integrations. - Features: 250+ pre-built tool integrations, Managed OAuth and authentication, Works with any agent framework, GitHub, Slack, Gmail, Jira, and more, Custom tool creation, Execution environment management, Action filtering and selection, Entity-based user management Tool integration platform that gives AI agents access to 250+ external tools and services with managed auth and execution. ### Mastra - URL: https://subagent.developersdigest.tech/frameworks#mastra - GitHub: https://github.com/mastra-ai/mastra - npm: @mastra/core - Category: Full-Stack - Language: TypeScript - Stars: 22k+ - Creator: Mastra - Best for: TypeScript developers wanting an opinionated, batteries-included agent framework with workflows and RAG. - Features: TypeScript-first design, Declarative workflow engine, Built-in RAG with vector stores, Agent memory and context, Tool calling with type safety, Evaluation and testing framework, Multi-provider LLM support, Syncs and integrations layer TypeScript-first agent framework with built-in workflows, RAG, integrations, and an evaluation system for AI applications. ### Pydantic AI - URL: https://subagent.developersdigest.tech/frameworks#pydantic-ai - GitHub: https://github.com/pydantic/pydantic-ai - pypi: pydantic-ai - Category: SDK - Language: Python - Stars: 15k+ - Creator: Pydantic - Best for: Python developers who value type safety and want Pydantic's validation power applied to agent development. - Features: Pydantic-based structured outputs, Type-safe tool definitions, Dependency injection for tools, Multi-model support, Streaming with validation, Result validators, Logfire integration for observability, System prompt templating Type-safe Python agent framework built on Pydantic, bringing validation, structured outputs, and dependency injection to AI agents. ### BeeAI Framework - URL: https://subagent.developersdigest.tech/frameworks#beeai-framework - GitHub: https://github.com/i-am-bee/beeai-framework - npm: beeai-framework - Category: SDK - Language: Both - Stars: 3k+ - Creator: IBM - Best for: Developers wanting a lightweight, IBM-backed agent framework with production observability. - Features: Python and TypeScript agents, ReAct agent pattern, Built-in tool library, Memory and context management, Serialization and persistence, Streaming support, OpenTelemetry observability, Custom tool creation Open-source framework for building production-ready AI agents in Python and TypeScript with tool use and memory. ### smolagents - URL: https://subagent.developersdigest.tech/frameworks#smolagents - GitHub: https://github.com/huggingface/smolagents - pypi: smolagents - Category: SDK - Language: Python - Stars: 25k+ - Creator: Hugging Face - Best for: Developers wanting a simple, code-first agent framework that leverages the Hugging Face ecosystem. - Features: Code-based agent execution, Minimal API surface, Multi-step reasoning, Tool calling via code generation, Hugging Face Hub integration, Multi-agent orchestration, Gradio UI for agent interaction, MCP server support Minimal Python agent framework by Hugging Face focused on code-based agents that write and execute Python to solve tasks. ### Semantic Kernel - URL: https://subagent.developersdigest.tech/frameworks#semantic-kernel - GitHub: https://github.com/microsoft/semantic-kernel - npm: semantic-kernel - pypi: semantic-kernel - Category: Orchestration - Language: Both - Stars: 27k+ - Creator: Microsoft - Best for: Enterprise teams integrating AI into existing .NET or Python applications with governance requirements. - Features: Enterprise-grade AI orchestration, Plugin architecture for tools, Planner for multi-step execution, Memory and vector store integration, Multi-provider LLM connectors, C#, Python, and Java SDKs, Prompt template engine, Responsible AI guardrails Microsoft enterprise AI orchestration SDK for integrating LLMs into conventional applications with plugins and planners. ### Haystack - URL: https://subagent.developersdigest.tech/frameworks#haystack - GitHub: https://github.com/deepset-ai/haystack - pypi: haystack-ai - Category: Full-Stack - Language: Python - Stars: 24k+ - Creator: deepset - Best for: Teams building production RAG systems and search pipelines with a modular, composable architecture. - Features: Modular pipeline architecture, RAG pipeline building, Document stores and retrievers, Custom component creation, Multi-provider LLM support, Evaluation framework, Pipeline serialization (YAML), deepset Cloud deployment Modular NLP framework for building production-ready RAG pipelines, search systems, and question-answering applications. ### LlamaIndex - URL: https://subagent.developersdigest.tech/frameworks#llamaindex - GitHub: https://github.com/run-llama/llama_index - npm: llamaindex - pypi: llama-index - Category: Full-Stack - Language: Both - Stars: 47k+ - Creator: LlamaIndex - Best for: Teams building data-intensive LLM applications that need sophisticated retrieval and indexing over private data. - Features: Advanced data indexing and retrieval, Query engine abstraction, Agent and tool abstractions, Multi-document agents, Workflow engine for complex pipelines, 100+ data loader integrations, LlamaCloud managed service, Evaluation and observability Data framework for connecting LLMs to external data sources with advanced indexing, retrieval, and query engine capabilities. ### Agno - URL: https://subagent.developersdigest.tech/frameworks#agno - GitHub: https://github.com/agno-agi/agno - pypi: agno - Category: Full-Stack - Language: Python - Stars: 38k+ - Creator: Agno - Best for: Python developers wanting a batteries-included agent framework with memory, knowledge, and team coordination out of the box. - Features: Built-in memory and knowledge bases, Multi-agent teams with coordination, Tool calling with pre-built toolkits, Structured outputs and validation, RAG with vector store integration, Reasoning and chain-of-thought agents, Agent monitoring and debugging UI, Multi-provider LLM support Full-stack AI agent framework (formerly PhiData) for building agents with memory, tools, knowledge bases, and multi-agent teams. ### Agency Swarm - URL: https://subagent.developersdigest.tech/frameworks#agency-swarm - GitHub: https://github.com/VRSEN/agency-swarm - pypi: agency-swarm - Category: Multi-Agent - Language: Python - Stars: 4k+ - Creator: VRSEN - Best for: Developers building collaborative multi-agent systems with defined communication hierarchies and shared state. - Features: Customizable agent roles and instructions, Inter-agent communication flows, OpenAI Assistants API integration, Shared state management, Tool creation with type safety, Agency-level orchestration, Gradio UI for testing, Custom communication topologies Multi-agent orchestration framework for creating collaborative AI agent swarms with customizable roles and communication flows. ### Instructor - URL: https://subagent.developersdigest.tech/frameworks#instructor - GitHub: https://github.com/instructor-ai/instructor - npm: instructor - pypi: instructor - Category: SDK - Language: Both - Stars: 12k+ - Creator: Jason Liu - Best for: Teams that need reliable structured data extraction from LLMs with type-safe validation and retry logic. - Features: Structured output extraction with Pydantic, Automatic retries with validation, Streaming partial responses, Multi-provider support (OpenAI, Anthropic, etc.), Multimodal extraction from images, Batch processing for high volume, TypeScript and Python SDKs, Fine-tuning dataset generation Lightweight library for extracting structured outputs from LLMs using Pydantic models, with retries, validation, and streaming support. ### DSPy - URL: https://subagent.developersdigest.tech/frameworks#dspy - GitHub: https://github.com/stanfordnlp/dspy - pypi: dspy - Category: Orchestration - Language: Python - Stars: 32k+ - Creator: Stanford NLP - Best for: Research teams and ML engineers who want to optimize LLM pipelines programmatically instead of hand-tuning prompts. - Features: Declarative language model programming, Automatic prompt optimization, Composable modules (ChainOfThought, ReAct, etc.), Built-in evaluation and metrics, Few-shot example bootstrapping, Multi-stage pipeline composition, Assertions and constraints, Retrieval-augmented modules Stanford framework for programming - not prompting - language models using composable modules, optimizers, and automated prompt tuning. ### Outlines - URL: https://subagent.developersdigest.tech/frameworks#outlines - GitHub: https://github.com/dottxt-ai/outlines - pypi: outlines - Category: SDK - Language: Python - Stars: 13k+ - Creator: .txt - Best for: Teams running local models that need guaranteed structured output conforming to schemas, regex, or grammars. - Features: Regex-guided text generation, JSON schema-constrained output, Context-free grammar support, Multiple choice generation, Works with local models (transformers, llama.cpp), Type-safe output guarantees, Pydantic model integration, Fast generation with compiled FSMs Structured text generation library that guarantees LLM outputs match a given format using regex patterns, JSON schemas, or grammars. ## MCP Servers (25) ### Filesystem - URL: https://subagent.developersdigest.tech/mcp#filesystem - GitHub: https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem - npm: @modelcontextprotocol/server-filesystem - Category: Files - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-filesystem /path/to/dir` - Tools: read_file, write_file, list_directory, create_directory, move_file, search_files, get_file_info, read_multiple_files Secure file system access with configurable allowed directories. Read, write, search, and manage files and directories. ### GitHub - URL: https://subagent.developersdigest.tech/mcp#github - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/github - npm: @modelcontextprotocol/server-github - Category: DevTools - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-github` - Tools: create_repository, search_repositories, get_file_contents, create_issue, create_pull_request, list_commits, list_branches, push_files Interact with GitHub repositories, issues, pull requests, branches, and files through the GitHub API. ### PostgreSQL - URL: https://subagent.developersdigest.tech/mcp#postgres - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/postgres - npm: @modelcontextprotocol/server-postgres - Category: Database - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-postgres postgresql://localhost/mydb` - Tools: query, list_tables, describe_table Connect to PostgreSQL databases to query tables, inspect schemas, and run read-only SQL queries. ### Slack - URL: https://subagent.developersdigest.tech/mcp#slack - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/slack - npm: @modelcontextprotocol/server-slack - Category: Communication - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-slack` - Tools: list_channels, post_message, reply_to_thread, get_channel_history, search_messages, get_users, get_thread_replies Interact with Slack workspaces to read channels, post messages, manage conversations, and search message history. ### Brave Search - URL: https://subagent.developersdigest.tech/mcp#brave-search - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/brave-search - npm: @modelcontextprotocol/server-brave-search - Category: Search - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-brave-search` - Tools: brave_web_search, brave_local_search Web and local search using the Brave Search API with support for web results, news, and local business listings. ### Puppeteer - URL: https://subagent.developersdigest.tech/mcp#puppeteer - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/puppeteer - npm: @modelcontextprotocol/server-puppeteer - Category: Browser - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-puppeteer` - Tools: navigate, screenshot, click, fill, evaluate, get_content Browser automation and web scraping using Puppeteer. Navigate pages, take screenshots, click elements, and extract content. ### SQLite - URL: https://subagent.developersdigest.tech/mcp#sqlite - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/sqlite - Category: Database - Creator: Anthropic - Install: `uvx mcp-server-sqlite --db-path /path/to/db.sqlite` - Tools: read_query, write_query, create_table, list_tables, describe-table, append_insight Query and manage SQLite databases with SQL execution, schema inspection, and data analysis capabilities. ### Memory - URL: https://subagent.developersdigest.tech/mcp#memory - GitHub: https://github.com/modelcontextprotocol/servers/tree/main/src/memory - npm: @modelcontextprotocol/server-memory - Category: AI - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-memory` - Tools: create_entities, create_relations, search_nodes, open_nodes, delete_entities, delete_relations Persistent knowledge graph memory system for maintaining context across conversations using entities and relations. ### Sequential Thinking - URL: https://subagent.developersdigest.tech/mcp#sequential-thinking - GitHub: https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking - npm: @modelcontextprotocol/server-sequential-thinking - Category: AI - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-sequential-thinking` - Tools: sequentialthinking Structured multi-step reasoning server that helps break down complex problems into sequential thought steps. ### Fetch - URL: https://subagent.developersdigest.tech/mcp#fetch - GitHub: https://github.com/modelcontextprotocol/servers/tree/main/src/fetch - Category: Search - Creator: Anthropic - Install: `uvx mcp-server-fetch` - Tools: fetch Fetch and convert web content to markdown for easy consumption. Handles HTML pages, PDFs, and other web resources. ### Firecrawl - URL: https://subagent.developersdigest.tech/mcp#firecrawl - GitHub: https://github.com/firecrawl/firecrawl-mcp-server - npm: firecrawl-mcp - Category: Search - Creator: Firecrawl - Install: `npx firecrawl-mcp` - Tools: firecrawl_scrape, firecrawl_crawl, firecrawl_search, firecrawl_extract, firecrawl_map Web scraping and crawling with AI-powered content extraction. Scrape single pages, crawl entire sites, and extract structured data. ### Playwright - URL: https://subagent.developersdigest.tech/mcp#playwright - GitHub: https://github.com/microsoft/playwright-mcp - npm: @playwright/mcp - Category: Browser - Creator: Microsoft - Install: `npx @playwright/mcp` - Tools: browser_navigate, browser_click, browser_fill_form, browser_snapshot, browser_take_screenshot, browser_evaluate, browser_tabs, browser_type, browser_hover, browser_press_key, browser_select_option, browser_wait_for Browser automation using Playwright with support for Chrome, Firefox, and WebKit. Includes accessibility tree snapshots and screenshots. ### Linear - URL: https://subagent.developersdigest.tech/mcp#linear - GitHub: https://github.com/dvcrn/mcp-server-linear - npm: linear-mcp - Category: DevTools - Creator: Community - Install: `npx linear-mcp` - Tools: create_issue, search_issues, get_issue, update_issue, list_projects, list_teams Interact with Linear for project management - create and search issues, manage projects, and track engineering workflows. ### Notion - URL: https://subagent.developersdigest.tech/mcp#notion - GitHub: https://github.com/makenotion/notion-mcp-server - npm: @notionhq/notion-mcp-server - Category: Data - Creator: Notion - Install: `npx @notionhq/notion-mcp-server` - Tools: search, get_page, create_page, update_page, query_database, create_database, get_block_children, append_block_children, create_comment Read, search, and manage Notion workspaces including pages, databases, comments, and content blocks. ### Obsidian - URL: https://subagent.developersdigest.tech/mcp#obsidian - GitHub: https://github.com/StevenStavrakis/obsidian-mcp - npm: obsidian-mcp - Category: Files - Creator: Community - Install: `npx obsidian-mcp /path/to/vault` - Tools: read_note, create_note, edit_note, search_vault, list_notes, get_tags Access and manage Obsidian vaults - read, create, search, and edit markdown notes with frontmatter support. ### Stripe - URL: https://subagent.developersdigest.tech/mcp#stripe - GitHub: https://github.com/stripe/ai - npm: @stripe/mcp - Category: Data - Creator: Stripe - Install: `npx @stripe/mcp` - Tools: create_customer, create_payment_link, list_products, create_invoice, list_subscriptions, get_balance Interact with the Stripe API for payments, subscriptions, customers, and billing operations. ### Supabase - URL: https://subagent.developersdigest.tech/mcp#supabase - GitHub: https://github.com/supabase-community/supabase-mcp - npm: @supabase/mcp-server-supabase - Category: Database - Creator: Supabase - Install: `npx @supabase/mcp-server-supabase` - Tools: list_projects, execute_sql, apply_migration, list_tables, get_logs, list_extensions, list_edge_functions Manage Supabase projects with database queries, auth management, storage operations, and edge function deployment. ### Neon - URL: https://subagent.developersdigest.tech/mcp#neon - GitHub: https://github.com/neondatabase/mcp-server-neon - npm: @neondatabase/mcp-server-neon - Category: Database - Creator: Neon - Install: `npx @neondatabase/mcp-server-neon` - Tools: list_projects, create_branch, run_sql, get_connection_string, list_branches, describe_table_schema Manage Neon serverless Postgres databases - create branches, run queries, manage schemas, and handle migrations. ### Sentry - URL: https://subagent.developersdigest.tech/mcp#sentry - GitHub: https://github.com/getsentry/sentry-mcp - npm: @sentry/mcp-server - Category: DevTools - Creator: Sentry - Install: `npx @sentry/mcp-server` - Tools: search_issues, get_issue_details, get_event, list_projects, get_issue_events Access Sentry error tracking data - search issues, view stack traces, and analyze error trends. ### Cloudflare - URL: https://subagent.developersdigest.tech/mcp#cloudflare - GitHub: https://github.com/cloudflare/mcp-server-cloudflare - npm: @cloudflare/mcp-server-cloudflare - Category: DevTools - Creator: Cloudflare - Install: `npx @cloudflare/mcp-server-cloudflare` - Tools: worker_list, worker_get, worker_put, kv_get, kv_put, r2_list, d1_query Manage Cloudflare Workers, KV namespaces, R2 buckets, and D1 databases through the Cloudflare API. ### Google Maps - URL: https://subagent.developersdigest.tech/mcp#google-maps - GitHub: https://github.com/modelcontextprotocol/servers-archived/tree/main/src/google-maps - npm: @modelcontextprotocol/server-google-maps - Category: Data - Creator: Anthropic - Install: `npx @modelcontextprotocol/server-google-maps` - Tools: geocode, reverse_geocode, search_places, get_directions, get_elevation, distance_matrix Access Google Maps APIs for geocoding, directions, place search, elevation data, and distance calculations. ### Exa - URL: https://subagent.developersdigest.tech/mcp#exa - GitHub: https://github.com/exa-labs/exa-mcp-server - npm: exa-mcp-server - Category: Search - Creator: Exa - Install: `npx exa-mcp-server` - Tools: web_search_exa, web_search_advanced_exa, get_code_context_exa, crawling_exa, company_research_exa, deep_search_exa Search the web and find similar content using Exa's AI-powered search API. Supports web search, code context lookup, company research, and deep research. ### Tavily - URL: https://subagent.developersdigest.tech/mcp#tavily - GitHub: https://github.com/tavily-ai/tavily-mcp - npm: tavily-mcp - Category: Search - Creator: Tavily - Install: `npx tavily-mcp` - Tools: tavily_search, tavily_extract, tavily_crawl, tavily_map, tavily_research AI-optimized web search, content extraction, site mapping, and crawling via the Tavily API. Supports configurable search depth, domain filtering, and comprehensive research. ### E2B - URL: https://subagent.developersdigest.tech/mcp#e2b - GitHub: https://github.com/e2b-dev/mcp-server - npm: @e2b/mcp-server - Category: DevTools - Creator: E2B - Install: `npx @e2b/mcp-server` - Tools: run_code Run code in secure cloud sandboxes via E2B. Execute Python, JavaScript, and other languages in isolated environments with code interpreter capabilities. ### Resend - URL: https://subagent.developersdigest.tech/mcp#resend - GitHub: https://github.com/resend/resend-mcp - npm: resend-mcp - Category: Communication - Creator: Resend - Install: `npx resend-mcp` - Tools: send-email, list-emails, get-email, send-batch-emails, create-contact, list-contacts, create-domain, list-domains Send and manage emails via the Resend platform. Supports transactional emails, batch sending, contacts, domains, broadcasts, and inbound email reading. ## Patterns (12) ### Swarm - URL: https://subagent.developersdigest.tech/patterns#swarm - Agents: 3-50+ - Complexity: Low - Best for: Parallel independent tasks like batch classification, multi-file analysis, bulk content generation, or processing large datasets. - Tradeoffs: High throughput but no inter-agent communication during execution. Workers cannot share context or build on each other's results. Resource usage scales linearly with worker count. - Frameworks: LangGraph, CrewAI, AutoGen, Mastra, Claude Agent SDK Fan out identical tasks to N worker agents running in parallel, then merge all results into a single output. How it works: A coordinator splits a task into independent subtasks and dispatches them to worker agents simultaneously. Each worker processes its subtask independently. When all workers complete, results are collected and merged into a unified response. ### Pipeline - URL: https://subagent.developersdigest.tech/patterns#pipeline - Agents: 2-6 - Complexity: Low - Best for: Multi-stage processing like research → draft → edit → publish, or extract → transform → load data pipelines. - Tradeoffs: Clear separation of concerns and predictable flow, but total latency is the sum of all stages. A failure at any stage blocks downstream agents. - Frameworks: LangGraph, CrewAI, Haystack, Mastra, Vercel AI SDK Sequential handoff between specialized agents where each stage transforms or enriches the output before passing it forward. How it works: Agents are arranged in a chain. Agent 1 produces output that becomes Agent 2's input, and so on. Each agent specializes in one transformation step. The final agent produces the end result. ### Supervisor - URL: https://subagent.developersdigest.tech/patterns#supervisor - Agents: 3-8 - Complexity: Medium - Best for: Complex projects requiring multiple skills — like building a feature that needs research, coding, testing, and documentation. - Tradeoffs: Flexible and adaptive, but the supervisor is a single point of failure. Quality depends heavily on the supervisor's ability to decompose tasks and evaluate results. - Frameworks: LangGraph, CrewAI, AutoGen, OpenAI Agents SDK, Semantic Kernel One orchestrator agent delegates tasks to specialist agents, reviews their work, and synthesizes a final result. How it works: The supervisor receives a task, breaks it down, and routes subtasks to the most appropriate specialist agents. It reviews each specialist's output, may request revisions, and compiles the final deliverable. ### Map-Reduce - URL: https://subagent.developersdigest.tech/patterns#map-reduce - Agents: 3-20+ - Complexity: Medium - Best for: Large-scale data processing, document summarization, multi-source analysis, and any task where divide-and-conquer applies. - Tradeoffs: Excellent for large inputs but the reduce step can become a bottleneck. Requires that the task be decomposable into independent chunks. - Frameworks: LangGraph, LangChain, AutoGen, Mastra, LlamaIndex Parallel map step processes chunks independently, then a reduce step aggregates and synthesizes all partial results. How it works: Input is split into chunks. The map phase sends each chunk to a worker agent for parallel processing. The reduce phase takes all mapped outputs and synthesizes them into a single coherent result, resolving conflicts and extracting patterns. ### Debate - URL: https://subagent.developersdigest.tech/patterns#debate - Agents: 3-5 - Complexity: Medium - Best for: Decision-making, strategy selection, code architecture choices, and any task where exploring multiple approaches improves quality. - Tradeoffs: Produces higher-quality outputs through adversarial evaluation, but uses 3-5x more tokens than a single agent. Not suitable for straightforward tasks. - Frameworks: AutoGen, LangGraph, CrewAI, Pydantic AI, smolagents Multiple agents propose competing solutions, then a critic agent evaluates arguments and selects or synthesizes the best approach. How it works: Proposer agents independently generate solutions to the same problem. A critic agent reviews all proposals, identifies strengths and weaknesses, and either selects the best one or synthesizes a superior solution from the best elements of each. ### Hierarchical - URL: https://subagent.developersdigest.tech/patterns#hierarchical - Agents: 7-50+ - Complexity: High - Best for: Large-scale orchestration with many agents, enterprise workflows, and projects that naturally decompose into teams and sub-teams. - Tradeoffs: Scales to many agents and complex projects, but introduces coordination overhead and latency from multiple management layers. Harder to debug when things go wrong. - Frameworks: LangGraph, CrewAI, AutoGen, Semantic Kernel, OpenAI Agents SDK Tree of supervisors and workers where top-level managers delegate to mid-level coordinators who manage specialized worker teams. How it works: A top-level supervisor breaks a large objective into domains. Each domain is assigned to a mid-level supervisor who further decomposes it into tasks for worker agents. Results flow back up the tree, with each level aggregating and quality-checking. ### Router - URL: https://subagent.developersdigest.tech/patterns#router - Agents: 3-10 - Complexity: Low - Best for: Heterogeneous workloads where different tasks require different models, tools, or expertise — like a support system handling billing, technical, and account queries. - Tradeoffs: Efficient routing reduces cost and improves quality, but accuracy depends on the classifier. Misrouted tasks can produce poor results. Adding new task types requires updating the router. - Frameworks: OpenAI Agents SDK, LangGraph, Vercel AI SDK, Semantic Kernel, Mastra A classifier agent analyzes incoming tasks and routes each one to the most appropriate specialist agent for handling. How it works: The router agent receives a task, classifies its type or domain, and forwards it to the matching specialist. Each specialist handles only the task types it excels at. This ensures optimal agent selection without wasting tokens on mismatched capabilities. ### Reflection - URL: https://subagent.developersdigest.tech/patterns#reflection - Agents: 1-2 - Complexity: Low - Best for: Quality-sensitive outputs like production code, published writing, complex analysis, and any task where iterative refinement significantly improves results. - Tradeoffs: Produces noticeably higher quality output, but each iteration adds latency and token cost. Requires well-defined quality criteria to avoid infinite loops. - Frameworks: LangGraph, AutoGen, CrewAI, Pydantic AI, Claude Agent SDK An agent generates output, then a critic (or the same agent) reviews and improves it iteratively until quality criteria are met. How it works: The generator agent produces an initial output. A critic agent (or the generator in self-critique mode) evaluates the output against quality criteria and provides specific feedback. The generator revises based on feedback. This loop repeats until the output meets the quality bar or a max iteration count is reached. ### Consensus - URL: https://subagent.developersdigest.tech/patterns#consensus - Agents: 3-7 - Complexity: Medium - Best for: High-stakes decisions, code review, content quality gates, and any task where independent perspectives reduce bias and catch errors. - Tradeoffs: Dramatically improves output quality through redundancy, but multiplies token cost by the number of agents. The mediator must be capable enough to judge solution quality accurately. - Frameworks: CrewAI, LangGraph, AutoGen Multiple agents independently solve the same problem, then a mediator picks the best solution or synthesizes a combined answer. Like best-of-N sampling at the agent level. How it works: N agents receive the same task and work on it independently with no communication. Each produces a complete solution. A mediator agent then reviews all solutions, scores them against criteria, and either selects the strongest one or synthesizes a superior answer by combining the best elements from each. ### Blackboard - URL: https://subagent.developersdigest.tech/patterns#blackboard - Agents: 3-10 - Complexity: High - Best for: Complex problem solving, collaborative analysis, research synthesis, and tasks where the solution emerges from incremental contributions by different specialists. - Tradeoffs: Highly flexible and supports emergent problem solving, but requires careful state management. Agents may conflict or overwrite each other's contributions. Harder to predict execution order and debug. - Frameworks: LangGraph, Semantic Kernel Agents share a common knowledge base (the blackboard) and contribute partial solutions. Each agent watches the blackboard and acts when it can contribute. How it works: A shared data store (the blackboard) holds the current problem state and partial solutions. Specialist agents monitor the blackboard and activate when they detect something they can contribute to. Each agent reads the current state, adds its partial solution, and writes back. A controller checks completion criteria after each update. ### Guardrail - URL: https://subagent.developersdigest.tech/patterns#guardrail - Agents: 2 - Complexity: Low - Best for: Safety-critical applications, content moderation, code validation, regulatory compliance, and any task where unchecked output poses risk. - Tradeoffs: Adds a reliable safety layer with minimal complexity, but introduces latency for every output. The guardrail agent must be fast and accurate - false positives block good output, false negatives defeat the purpose. - Frameworks: OpenAI Agents SDK, Claude Agent SDK, LangGraph A primary agent does the work while a secondary agent monitors and validates output in real-time. The guardrail agent can block, modify, or flag unsafe or incorrect outputs. How it works: The primary agent generates output as normal. Before the output is delivered, a guardrail agent intercepts and evaluates it against safety rules, quality standards, or business constraints. If the output passes, it goes through. If it fails, the guardrail either blocks it, requests a revision, or modifies it directly. ### Tool Augmented - URL: https://subagent.developersdigest.tech/patterns#tool-augmented - Agents: 1 - Complexity: Low - Best for: General-purpose automation, developer workflows, Claude Code-style usage, and tasks where a single smart agent with the right tools outperforms a team of agents. - Tradeoffs: Simple architecture with no coordination overhead, but limited by the agent's ability to plan and use tools correctly. Performance ceiling is bounded by a single model's reasoning capacity. - Frameworks: Claude Code, Vercel AI SDK, LangChain A single agent with access to a rich tool ecosystem - MCP servers, APIs, file system, databases. Not multi-agent per se, but the tools act as specialized capabilities that extend the agent's reach. How it works: One agent receives a task and has access to a registry of tools (via MCP, function calling, or plugin systems). It plans a sequence of tool calls, executes them, observes results, and iterates until the task is complete. The agent decides which tools to use, in what order, and how to combine their outputs. ## Skills (17) ### Frontend Design - URL: https://subagent.developersdigest.tech/skills#frontend-design - Repo: https://github.com/anthropics/skills/tree/main/skills/frontend-design - Category: Creative - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: ui, design, frontend, css, components Create distinctive, production-grade frontend interfaces with high design quality. Generates creative, polished web components, pages, and applications that avoid generic AI aesthetics. ### Canvas Design - URL: https://subagent.developersdigest.tech/skills#canvas-design - Repo: https://github.com/anthropics/skills/tree/main/skills/canvas-design - Category: Creative - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: canvas, png, pdf, visualization, graphics, art Create beautiful visual art in PNG and PDF documents using design philosophy. Use for posters, art pieces, designs, or other static visual content with original creative direction. ### Algorithmic Art - URL: https://subagent.developersdigest.tech/skills#algorithmic-art - Repo: https://github.com/anthropics/skills/tree/main/skills/algorithmic-art - Category: Creative - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: generative, p5js, art, algorithms, particles Create algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Generates flow fields, particle systems, and generative art from computational aesthetic philosophies. ### Theme Factory - URL: https://subagent.developersdigest.tech/skills#theme-factory - Repo: https://github.com/anthropics/skills/tree/main/skills/theme-factory - Category: Creative - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: theme, design-system, colors, typography, styling Apply consistent, professional styling to artifacts like slides, docs, reports, and landing pages. Includes 10 pre-set themes with curated color palettes and font pairings, or generate new themes on the fly. ### Brand Guidelines - URL: https://subagent.developersdigest.tech/skills#brand-guidelines - Repo: https://github.com/anthropics/skills/tree/main/skills/brand-guidelines - Category: Creative - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: brand, style-guide, identity, guidelines, colors Apply official brand colors and typography to artifacts. Use when brand colors, style guidelines, visual formatting, or company design standards need to be applied consistently. ### Slack GIF Creator - URL: https://subagent.developersdigest.tech/skills#slack-gif-creator - Repo: https://github.com/anthropics/skills/tree/main/skills/slack-gif-creator - Category: Creative - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: gif, animation, slack, emoji Create animated GIFs optimized for Slack with proper dimensions, frame rates, and file size constraints. Provides validation tools and animation concepts for emoji and message GIFs. ### Claude API - URL: https://subagent.developersdigest.tech/skills#claude-api - Repo: https://github.com/anthropics/skills/tree/main/skills/claude-api - Category: Development - Install: `/plugin install claude-api@anthropic-agent-skills` - Tags: api, anthropic, sdk, claude, agents Build apps with the Claude API and Anthropic SDKs. Covers the Messages API, Agent SDK, streaming, tool use, and best practices for Python and TypeScript. ### MCP Builder - URL: https://subagent.developersdigest.tech/skills#mcp-builder - Repo: https://github.com/anthropics/skills/tree/main/skills/mcp-builder - Category: Development - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: mcp, server, protocol, tools, integrations Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services. Supports Python (FastMCP) and Node/TypeScript (MCP SDK). ### Web Artifacts Builder - URL: https://subagent.developersdigest.tech/skills#web-artifacts-builder - Repo: https://github.com/anthropics/skills/tree/main/skills/web-artifacts-builder - Category: Development - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: react, artifacts, tailwind, shadcn, html Build elaborate, multi-component Claude.ai HTML artifacts using React, Tailwind CSS, and shadcn/ui. Handles complex artifacts requiring state management, routing, and bundling into single HTML files. ### Skill Creator - URL: https://subagent.developersdigest.tech/skills#skill-creator - Repo: https://github.com/anthropics/skills/tree/main/skills/skill-creator - Category: Development - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: skills, meta, evals, benchmarks, optimization Create new skills, modify and improve existing skills, and measure skill performance. Supports iterative development with eval testing, benchmarking, and description optimization for better triggering. ### Web App Testing - URL: https://subagent.developersdigest.tech/skills#webapp-testing - Repo: https://github.com/anthropics/skills/tree/main/skills/webapp-testing - Category: Code Quality - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: testing, playwright, browser, screenshots, qa Test local web applications using Playwright. Verify frontend functionality, debug UI behavior, capture browser screenshots, and view browser logs with automated test scripts. ### PDF - URL: https://subagent.developersdigest.tech/skills#pdf - Repo: https://github.com/anthropics/skills/tree/main/skills/pdf - Category: Documents - Install: `/plugin install document-skills@anthropic-agent-skills` - Tags: pdf, forms, ocr, merge, extract Read, create, merge, split, rotate, watermark, encrypt, and fill PDF forms. Includes OCR for scanned PDFs and image extraction using Python libraries and command-line tools. ### Word Documents - URL: https://subagent.developersdigest.tech/skills#docx - Repo: https://github.com/anthropics/skills/tree/main/skills/docx - Category: Documents - Install: `/plugin install document-skills@anthropic-agent-skills` - Tags: word, docx, documents, formatting, reports Create, read, edit, and manipulate Word documents (.docx). Supports tables of contents, headings, page numbers, letterheads, tracked changes, comments, and find-and-replace. ### PowerPoint - URL: https://subagent.developersdigest.tech/skills#pptx - Repo: https://github.com/anthropics/skills/tree/main/skills/pptx - Category: Documents - Install: `/plugin install document-skills@anthropic-agent-skills` - Tags: powerpoint, pptx, slides, presentations, decks Create, read, edit, and manipulate PowerPoint presentations (.pptx). Build slide decks, pitch decks, and presentations with templates, layouts, speaker notes, and comments. ### Excel Spreadsheets - URL: https://subagent.developersdigest.tech/skills#xlsx - Repo: https://github.com/anthropics/skills/tree/main/skills/xlsx - Category: Documents - Install: `/plugin install document-skills@anthropic-agent-skills` - Tags: excel, xlsx, spreadsheets, formulas, data Create, read, edit, and process Excel spreadsheets (.xlsx). Handles formulas, formatting, charting, data cleaning, and conversion between tabular file formats. ### Doc Co-Authoring - URL: https://subagent.developersdigest.tech/skills#doc-coauthoring - Repo: https://github.com/anthropics/skills/tree/main/skills/doc-coauthoring - Category: Documents - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: documentation, writing, specs, proposals, collaboration Structured workflow for co-authoring documentation through three stages: context gathering, refinement and structure, and reader testing. For proposals, technical specs, decision docs, and more. ### Internal Comms - URL: https://subagent.developersdigest.tech/skills#internal-comms - Repo: https://github.com/anthropics/skills/tree/main/skills/internal-comms - Category: Productivity - Install: `/plugin install example-skills@anthropic-agent-skills` - Tags: communications, reports, updates, newsletters, writing Write internal communications including status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, and project updates using standard company formats.