DeepSeek V4 Flash 0731

Jul 2026
Older versionReplace withDeepSeek V4.1 Flash 1017Newer versionMiMo v2.6 ProCheaper, as goodClaude Haiku 5.5Better
DeepSeek V4 Flash 0731
Compare in
on
$0.17per task?
$0.14 / $0.28per 1M in / out?
131 tok/sfirst token in 1.2 s?
1Mcontext

Rankingat max effort?

Retired models are left out of the current rankings.

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperMiMo v2.6 ProthinkingOpen+35%-24%$0.13Pass back reasoning_content+3BetterMiMo v2.6 FlashthinkingOpen+10%-64%$0.062Pass back reasoning_content+3CheaperGPT-6.1 Sollow+23%-25%$0.13Remove unsupported parameters+5CheaperClaude Haiku 5.5xhigh+20%-29%$0.12Remove sampling parameters+5CheaperGPT-6 Lunamax+11%-61%$0.068Function calling only with none · Chat Completions+3

Run it yourself

Fits onhardware needs not verified yet
Licencenot verified

History

4 versions
2530354045Jul 26Oct 26Release dateAA Intelligence Index 4.3.2V4 · AprV4 · JulV4.1 · SepV4.1 · Oct
2530354045Jul 26Oct 26Release dateAA Intelligence Index 4.3.2V4 · AprV4 · JulV4.1 · Sep

Benchmarks

all categories
Chat & Knowledge
Humanity's Last Exam38.6
AA Omniscience Index-14.3
SimpleQA Verified v133.6
CritPt (physics)16.6
AA Intelligence Index 4.3.234.3
AA Omniscience · Accuracy40.4
AA Omniscience · Hallucination Rate91.7
AA-LCR (long context) 1.179.7
GDP.pdf · AA all-pass11
MLCR-AA13.3
ARC-AGI-1 · public evaluation94.8
ARC-AGI-1 · semi-private89
ARC-AGI-2 · public evaluation63.9
ARC-AGI-2 · semi-private61.4
FrontierMath Tiers 1–3 (v2)57.5
FrontierMath Tier 4 (v2)24.4
Chess Puzzles32
Mystery Game Puzzles34
GPQA Diamond90.8
GPQA Diamond · Epoch91
OTIS Mock AIME 2024 202594.4
Coding
Terminal-Bench 4.012.1
SciCode50.3
Terminal-Bench 2.178.7
Agents & Tool Use
AutomationBench-AA54
GDPval-AA (work tasks) 2.146.7
τ³-Banking · AA39.4
AA-Briefcase 1.1 · Elo1,258.6

FAQ

How much does DeepSeek V4 Flash 0731 cost?›

$0.14 per 1M input tokens and $0.28 per 1M output tokens. A typical benchmark task costs about $0.17 at max effort.

Is DeepSeek V4 Flash 0731 free to use?›

There is no free API; the cheapest paid price is $0.14 per 1M input tokens. The weights are open, so you can run it yourself.

How can I use DeepSeek V4 Flash 0731?›

Through the DeepSeek API.

What is the context window of DeepSeek V4 Flash 0731?›

1M tokens.

When was DeepSeek V4 Flash 0731 released? Is there a newer version?›

Released 2026-07-31. It has been superseded by DeepSeek V4.1 Flash.

Is DeepSeek V4 Flash 0731 open source?›

Yes, the weights are open.

Can I run DeepSeek V4 Flash 0731 locally?›

The weights are open and can be self-hosted; hardware needs are not confirmed yet.

What is DeepSeek V4 Flash 0731?›

DeepSeek V4 Flash 0731 is an open-weights language model from DeepSeek, released Jul 2026, the successor to DeepSeek V4 Flash 0420.

What are the best alternatives to DeepSeek V4 Flash 0731?›

MiMo v2.6 Pro (cheaper), MiMo v2.6 Flash (better).