Back to Models

DeepSeek V4 Flash

![DeepSeek](https://artificialanalysis.ai/img/logos/deepseek_small.svg)DeepSeek
36/ 57

Intelligence Score

Intelligence

36

#59 of 212 (top 28%)

Speed

83 tok/s

#85 of 141

Price

$0.18

per M tokens

#26 cheapest of 142

Context Window

1M

tokens

Latency: 1ms (#45)

Score Breakdown

Quality Score60%
Speed Score11%
Cost Score98%

Similar Models

NVIDIA Nemotron 3 Super

![NVIDIA](https://artificialanalysis.ai/img/logos/nvidia_small.svg)NVIDIA

36intelligence

Qwen3.5 122B A10B

![Alibaba](https://artificialanalysis.ai/img/logos/alibaba_small.svg)Alibaba

36intelligence

Nova 2.0 Pro Preview (medium)

![Amazon](https://artificialanalysis.ai/img/logos/aws_small.svg)Amazon

36intelligence

MiMo-V2.5-Pro

![Xiaomi](https://artificialanalysis.ai/img/logos/xiaomi_small.svg)Xiaomi

36intelligence

Kimi K2.5

![Kimi](https://artificialanalysis.ai/_next/image?url=%2Fimg%2Flogos%2Fkimi_small.png&w=32&q=75)Kimi

37intelligence

Use This Model

Claude Code

claude --model deepseek-v4-flash-non-reasoning

API (cURL)

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -d '{
    "model": "deepseek-v4-flash-non-reasoning",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Vercel AI SDK

import { generateText } from "ai";
import { openrouter } from "@openrouter/ai-sdk-provider";

const { text } = await generateText({
  model: openrouter("deepseek-v4-flash-non-reasoning"),
  prompt: "Hello",
});
Back to Models