DeepSeek Flash
by DeepSeek · Model · Language & code models
strong
Best-value open-weights model: near-frontier work for a few dollars, though a step behind.
DeepSeek V4.1 Flash 1017 released October 17 as an update to V4.1 Flash; reviewer calls it the best open model
AccessOpen weightsNeeds~510 GBDownloadabout 510 GB full weights; 1-bit GGUF is 233 GBLicenseMIT
Why we say so
Arena ranks V4.1 Flash second to third among open-weights models on web development and agents, behind Kimi K3.1. AI Search ran it through multi-hour projects for about $6 and found it roughly 80-90% as good as GPT-6 Astra and Opus, with it alone finding the hidden frog in an image test. Much of the benchmark comparison against GLM and Kimi comes from DeepSeek itself, and one HN commenter prefers it over a local Qwen for real work.
Versions
| Version | Released | Rank |
|---|---|---|
| DeepSeek V4.1 FlashLATEST | Sep 9 | #2 web development (open weights) |
| DeepSeek V4 Flash | Jul 31 |
Evidence
- Nov 19A swarm of 1,000 agents wrote a 630k-line zero-dependency rule-based chatbot in a week (~8k merged PRs), reported on par with SmolLM-360M
- Nov 15#2 on Arena webdev among open weights (1659)
- Nov 11#2 on Arena webdev among open weights (1664)
- Oct 23Intelligence-versus-cost-per-task chart puts it more cost-efficient than the top GPT and Claude models
- Oct 23Beats GLM 5.3 and Kimi K3 across the shown benchmarks, including agentic coding and general knowledge
Show all 31
- Oct 23Benchmark performance roughly on par with GPT Sol, matching the reviewer's experience; below GPT-6 Astra
- Oct 23Ranks third on KernelBench CUDA, below the latest Claude models but ahead of Gemini 4 Argon, Claude Fable and GPT-6 Astra
- Oct 23Unreal Engine game with Blender assets: playable but critic scored only 5.5/10 after 16 rounds, about 80% of frontier models
- Oct 2320-minute Alzheimer's deep-research HTML report judged tied with GPT-6 Astra and much better than Opus 5.5
- Oct 23Medical deep-research report much better than Opus 5.5, which was lazy and low effort
- Oct 19Took #1 on Arena agent among open weights
- Oct 19#3 on Arena vision among open weights (1278)
- Oct 18Took #1 on Arena webdev among open weights
- Sep 27MiMo 2.6 is cheaper per task than DeepSeek V4.1 Flash
- Sep 18552B MoE with encoder-decoder design; open weights (~510 GB), cheap fast API; claims top open-model scores on several leaderboards
- Sep 18Ranked #1 open model on LiveBench and #1 on Vals Index; far cheaper per task than GPT-6 Astra
- Sep 18Global KV cache cut to 890 bytes/token, about 437x below DeepSeek V1 and ~4x below V4 Flash
- Sep 18Scores 74.2 on DeepSWE v1.1 versus GPT-6 Astra's 74, at lower cost and about 4x output speed
- Sep 13Tied with Kimi K3 and GLM 5.3 on Vals Index, behind on Artificial Analysis, but 3-7x faster via API
- Sep 9DeepSeek claims V4.1 Flash comprehensively beats V4 Pro across all key metrics
- Sep 9Found and fixed a flipped comparison operator in a Dockerized air-traffic dashboard without being told, then re-checked the live endpoint eight times
- Sep 9Reported at 300-400+ tokens per second and 98% of GPT-6 Astra's design benchmark score; unverified
- Sep 9Unverified third-party numbers put it against Opus 4.8 at max thinking, strong in coding and tool use
- Sep 9Preview not final: access expected to be removed around September 10 pending post-training; open weights hoped for with the finished version
- Aug 21The July V4 Flash build (0731) reportedly assumed it could see images and broke sessions trying to read screenshots
- Aug 4284B-parameter MoE runs fully locally on one Ryzen AI MAX+ 395 with 128GB unified memory, using a 95GB target plus 11GB draft
- Aug 4In 3D oil-rig test, built the most literal, detailed rig with half-submerged brief met exactly, but more engineering render than landing page
- Aug 2New 0731 Flash version released, same structure as the DSpark variant; about 10 points above the previous V4 Flash
- Aug 2Beats MiniMax M3 and the previous V4 Pro, one point below GLM 5.2 on the index shown
- Aug 2Beats or matches GLM 5.2 and Opus 4.8 on some agentic coding, software engineering and cybersecurity benchmarks
- Jul 31In one chained-bug full-stack repair task, it fixed all five bugs including a silent Redis cache-key mismatch; reviewer says Fable and the newest GPT failed it
Cost to run
Alternatives
More DeepSeek Flash
Built on DeepSeek Flash