DeepSeek ships V4.1-Flash-1017: stronger benchmarks, lower cost, now open-source
deepseek_ai · x · 2026-10-17
DeepSeek released DeepSeek-V4.1-Flash-1017, live on web, app, and API.
- Pareto improvement over the previous version: better benchmark scores, faster task completion, shorter outputs
- Optimized for DeepSeek Harness, Tencent CodeBuddy, WorkBuddy; Harness v0.2.1-rc.1+ recommended
- Benchmark charts: Terminal-Bench 4.0 at 42.3 (vs 31.2 prior), Automation-Bench 60.5, Apex Bench 72.1, beating Kimi K3, GLM-5.3, and GPT-5.6-Sol on several agent benchmarks
- DeepSWE v1.1 cost curve: 77% Pass@1 at roughly $0.2 per task, on the Pareto frontier
More from Models
- Lava lamp test: Gemini 4 nails it for $0.18 in 2 min vs Opus 5.5's $1.27 and 9.5 min — CodeByPoonam · 2026-10-17
- Gemini 4 Argon builds a Minecraft clone in 19 minutes using 6.3M tokens — CodeByPoonam · 2026-10-17
- Scaling01 criticizes Kimi K3 as undertrained: big but starving for tokens — scaling01 · 2026-10-17
- Unreleased Claude model "onigiri-eap-2" surfaces with four reasoning levels — mark_k · 2026-10-17
- OpenAI's internal data: model task time horizons double every 64 days — LeviTurk · 2026-10-17
- DeepSeek-V4.1-Flash-1017 released open-source under MIT license — deepseek_ai · 2026-10-17