DeepSeek Releases V4.1-Flash-1017, Opens Search and Fetch APIs to All Developers
DeepSeek launched its V4.1-Flash-1017 model and opened Search and Fetch APIs to all users, with free access through October 2026 for select tools.
Quick answer
What did DeepSeek announce with its V4.1-Flash-1017 release?
DeepSeek released DeepSeek-V4.1-Flash-1017, which the company says improves benchmark performance, speed, and output length over its predecessor. DeepSeek also opened its Search and Fetch APIs to all developers at $0.75 per 1,000 calls, with free use through certain tools until October 31, 2026. The model is open-sourced on Hugging Face and ModelScope.
Key takeaways
- DeepSeek released DeepSeek-V4.1-Flash-1017 on its website, app, and API, describing it as a Pareto improvement in performance and cost over the prior version.
- The company says the model is optimized for DeepSeek Harness, Tencent CodeBuddy, and WorkBuddy, and recommends upgrading to DeepSeek Harness v0.2.1-rc.1 or later.
- DeepSeek opened its in-house Search and Fetch APIs to all users at $0.75 per 1,000 calls per endpoint, with billing for server-side tool calls in DeepSeek Harness, Claude Code, and Codex starting October 31, 2026 at 16:00 UTC.
- The model weights are open-sourced and available on Hugging Face and ModelScope.
- DeepSeek showcased the model's project-building capabilities with examples including 3D modeling, a web-based racing game, and a Godot-engine recreation of Portal 2.
New model goes live across DeepSeek's platforms
DeepSeek has released DeepSeek-V4.1-Flash-1017, making it available immediately on the company's website, app, and API. DeepSeek says the model delivers stronger benchmark performance, faster task completion, and shorter outputs compared with its predecessor, which the company describes as a Pareto improvement in both performance and cost.
DeepSeek says the model is optimized for use with DeepSeek Harness, Tencent CodeBuddy, and WorkBuddy, and recommends developers upgrade to DeepSeek Harness version 0.2.1-rc.1 or later to get the best experience with the new model.
The company published a chart comparing Pass@1 scores against cost per task for the model on DeepSWE v1.1, tested using the mini-swe-v2 harness, though it did not provide the underlying figures in the announcement.
Built for end-to-end project generation
DeepSeek says the new model uses reinforcement learning with self-feedback to improve its ability to build complete projects end to end, ranging from game creation to 3D modeling. The company shared several examples it says demonstrate this capability: a 3D model of the Forbidden City, a web-based racing game, a recreation of Portal 2 built with the Godot engine, and a 3D scene called "Valley of Ages" depicting the evolution from prehistoric times to the modern era.
Search and Fetch APIs open to all
DeepSeek also announced that its Search and Fetch APIs, powered by its in-house search engine, are now open to all developers. The company says the search engine serves hundreds of millions of requests daily worldwide.
DeepSeek says calls made to these APIs through server-side tools in DeepSeek Harness, Claude Code, and Codex are currently free for DeepSeek API users, with billing set to begin on October 31, 2026 at 16:00 UTC.
Open-source release
DeepSeek has made DeepSeek-V4.1-Flash-1017 available as an open-source release. The model can be downloaded from Hugging Face and ModelScope.
Source: DeepSeek API Docs, "DeepSeek-V4.1-Flash-1017: Wish-Driven Programming Made Real," published October 17, 2026.
Frequently asked questions
- What is DeepSeek-V4.1-Flash-1017?
- It is DeepSeek's latest model release, which the company says delivers stronger benchmark performance, faster task completion, and shorter outputs than its previous version.
- What are DeepSeek's Search and Fetch APIs?
- They are APIs powered by DeepSeek's in-house search engine that let developers perform web search and fetch operations, now open to all users at $0.75 per 1,000 calls per endpoint.
- Is DeepSeek-V4.1-Flash-1017 free to use?
- The model itself is open-sourced on Hugging Face and ModelScope. Separately, Search and Fetch calls made through server-side tools in DeepSeek Harness, Claude Code, and Codex are free for DeepSeek API users until billing begins on October 31, 2026 at 16:00 UTC.
- What tools is the new model optimized for?
- DeepSeek says the model is optimized for DeepSeek Harness, Tencent CodeBuddy, and WorkBuddy, and recommends upgrading to DeepSeek Harness version 0.2.1-rc.1 or later for the best experience.