DeepSeek

DeepSeek V4 Flash

CTX1M

NewRetired

71Quality

Reasoning · Vision · Tools

Input$0.15/M

Output$0.60/M

October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding

Open weightsMIT1041/1044

Drag to turn

DeepSeekTextCommonRetired

DeepSeek V4 Flash

October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding

deepseek-v4-flash

Type: Text
Quality: 71 · Top 29% on the Epoch Capabilities Index
Context window: 1,000,000 tokens
Max output: 393,216 tokens
Input price: $0.15 / 1M tokens
Output price: $0.60 / 1M tokens
Cache read: $0.003 / 1M tokens
Cache write: —
Parameters: 291B (290,944,616,402)
Downloads: 800K in the last 30 days
Likes: 2.3K
Access to weights: Open download
Released: Oct 17, 2026
Knowledge cutoff: May 2025
Takes in: Text, Image
Puts out: Text
Reasoning: Yes
Tool calling: Yes
Structured output: Yes
File attachments: Yes
Access: Open weights
License: MIT
Rarity: Common
Set number: 1041/1044

Open DeepSeek docs Weights on Hugging Face Quality 71 Where to run it 66 Evolution 1/2 Meet its creature Battle MiniMax-M3.1-Flash

License read from the model’s Hugging Face card. Specs from models.dev, downloads from Hugging Face, last changed Oct 19, 2026. Official source: api-docs.deepseek.com.

How good is it?

Public leaderboard scores for DeepSeek V4 Flash, refreshed daily from Epoch AI and LMArena.

Top 29% of the 277 models on the Epoch Capabilities Index: #81.

Quality is the share of a leaderboard’s models this one beats, from 0 to 100. Chat models are rated on Epoch AI’s Capabilities Index when it lists them, otherwise on LMArena; image and video models on LMArena’s image and video arenas.

Leaderboards

Epoch Capabilities Index#81 of 277
One number fitted from dozens of hard benchmarks, from science questions to agentic coding.146.1 index
LMArena Text#92 of 416
People compare two anonymous answers and vote for the better one.1,432 rating
LMArena Text · Coding#88 of 411
The text arena’s coding prompts only.1,457 rating
LMArena Text · Math#107 of 400
The text arena’s math prompts only.1,426 rating
LMArena Text · Creative writing#91 of 414
The text arena’s creative writing prompts only.1,403 rating
LMArena WebDev#34 of 145
Each model builds a web app from the same prompt; people pick the better one.1,582 rating
Listed as DeepSeek-V4-Flash (High)

Scores from Epoch AI and LMArena, licensed CC BY 4.0. Boards list models under their own names, so a close match is shown with the name it was scored under. Credits

Where to run it

64 providers sell DeepSeek V4 Flash through an API, and 2 Hugging Face hosts serve it. Prices are per million tokens, cheapest first.

CheapestCrofAI$0.08 in · $0.10 out
Free to try12 providersAlibaba Token Plan, Alibaba Token Plan (China), InferX, …
From the labDeepSeek$0.15 in · $0.60 out
Fastest on Hugging FaceDeepInfra18 tokens/s

API providers

Provider	Input	Output	Context
Alibaba Token PlanFree`deepseek-v4-flash`	Free	Free	1M
Alibaba Token Plan (China)Free`deepseek-v4-flash`	Free	Free	1M
InferXFree`deepseek-v4-flash`	Free	Free	1M
KenariFree`deepseek-v4-flash`	Free	Free	1M
NaNFree`deepseek-v4-flash`	Free	Free	1M
OrcaRouterFree`deepseek/deepseek-v4-flash-free`	Free	Free	1M
PendraFree`deepseek-v4-flash`	Free	Free	1M
SCNet Token PlanFree`DeepSeek-V4-Flash`	Free	Free	1M
SenseNova (China)Free`deepseek-v4-flash`	Free	Free	1M
Umans AI Coding PlanFree`umans-deepseek-v4-flash-0731`	Free	Free	1.05M

Show 54 more providers

Provider	Input	Output	Context
UnoRouterFree`deepseek-v4-flash:free`	Free	Free	1M
Volcengine Ark Coding PlanFree`deepseek-v4-flash`	Free	Free	1M
CrofAICheapest paid`deepseek-v4-flash-0731`	$0.08	$0.10	1M
DevPass (LLM Gateway)`deepseek-v4-flash`	$0.076	$0.15	1.05M
LLM Gateway`runware/deepseek-v4-flash`	$0.076	$0.15	1.05M
Standard Compute`deepseek/deepseek-v4-flash`	$0.083	$0.17	1M
Deep Infra`deepseek-ai/DeepSeek-V4-Flash`	$0.09	$0.18	1.05M
Kilo Gateway`deepseek/deepseek-v4-flash`	$0.091	$0.18	1.05M
OpenRouter`deepseek/deepseek-v4-flash`	$0.091	$0.18	1.05M
TokenGo`deepseek/deepseek-v4-flash`	$0.098	$0.20	1M
Modelis`deepseek-v4-flash`	$0.098	$0.20	1M
Pioneer`deepseek-ai/DeepSeek-V4-Flash`	$0.10	$0.20	1M
GMI Cloud`deepseek-ai/DeepSeek-V4-Flash`	$0.11	$0.22	1.05M
routing.run`deepseek-v4-flash`	$0.11	$0.22	1M
ai&`deepseek-ai/deepseek-v4-flash`	$0.15	$0.25	1.05M
Vercel AI Gateway`deepseek/deepseek-v4-flash`	$0.13	$0.26	1M
Venice AI`deepseek-v4-flash`	$0.14	$0.28	1M
SiliconFlow`deepseek-ai/DeepSeek-V4-Flash`	$0.13	$0.28	1M
Abacus`deepseek-ai/DeepSeek-V4-Flash`	$0.14	$0.28	1M
AIHubMix`alicloud-deepseek-v4-flash`	$0.14	$0.28	1M
Alibaba (China)`deepseek-v4-flash`	$0.14	$0.28	1M
Ambient`deepseek/deepseek-v4-flash`	$0.14	$0.28	1.05M
AMD`DeepSeek-V4-Flash`	$0.14	$0.28	1.05M
Auriko`deepseek-v4-flash`	$0.14	$0.28	1M
ClinePass`cline-pass/deepseek-v4-flash`	$0.14	$0.28	1M
DigitalOcean`deepseek-4-flash`	$0.14	$0.28	1.05M
HPC-AI`deepseek/deepseek-v4-flash`	$0.14	$0.28	1.05M
Jalapeno Cloud`DeepSeek-V4-Flash`	$0.14	$0.28	1.05M
NanoGPT`deepseek/deepseek-v4-flash`	$0.14	$0.28	1.05M
Neuralwatt`deepseek-v4-flash`	$0.14	$0.28	1.05M
Nvidia`deepseek-ai/deepseek-v4-flash`	$0.14	$0.28	1.05M
OpenCode`deepseek-v4-flash`	$0.14	$0.28	1M
SiliconFlow (China)`deepseek-ai/DeepSeek-V4-Flash`	$0.14	$0.28	1M
Umans AI`umans-deepseek-v4-flash-0731`	$0.14	$0.28	1.05M
ZenMux`deepseek/deepseek-v4-flash`	$0.14	$0.28	1M
EBCloud`DeepSeek-V4-Flash`	$0.14	$0.29	1M
Vivgrid`deepseek-v4-flash`	$0.15	$0.30	1M
Moark`DeepSeek-V4-Flash`	$0.50	$0.50	1M
Azure`deepseek-v4-flash`	$0.19	$0.51	1M
Impossibl`deepseek/deepseek-v4-flash`	$0.19	$0.51	1M
Requesty`deepseek-v4-flash`	$0.28	$0.56	1.05M
CommandCode`deepseek/deepseek-v4-flash`	$0.15	$0.60	1M
DeepSeekOfficial`deepseek-v4-flash`	$0.15	$0.60	1M
Eden AI`deepseek/deepseek-v4-flash`	$0.15	$0.60	1M
above.dev`deepseek-v4-flash`	$0.17	$0.66	1M
Ollama Cloud`deepseek-v4-flash`	$0.22	$0.66	1.05M
Opper`deepseek-v4-flash`	$0.25	$0.66	1M
CrossModel`deepseek/deepseek-v4-flash`	$0.30	$1.20	1M
Merge Gateway`deepseek/deepseek-v4-flash`	$0.30	$1.20	1M
Ofox`deepseek/deepseek-v4-flash`	$0.44	$1.32	1M
Prime Inference`deepseek/deepseek-v4-flash`	$0.44	$1.32	1.05M
AI/ML API`deepseek/deepseek-chat`	$0.39	$1.56	1M
AgentRouter`deepseek-v4-flash`	—	—	1M
AnyAPI`deepseek/deepseek-v4-flash`	—	—	1M

Hugging Face Inference Providers

Host	Input	Output	Context	Speed	First token
DeepInfraFastestTools	$0.09	$0.18	1.05M	18 tok/s	3.1s
Featherless	—	—	—	—	—

From models.dev and Hugging Face, checked daily. Resellers’ listings can lag behind their price pages, so confirm with the provider before you commit.

Face off

Current text models from other labs at a similar output price.