AI BriefingDeepSeek Releases V4.1-Flash-1017 and Opens Search & Fetch APIs
Key point
The new model is open-sourced on Hugging Face and ModelScope, while Search & Fetch APIs are now available for $0.75 per 1,000 calls.
Details
DeepSeek has released DeepSeek-V4.1-Flash-1017, a model update described as a Pareto improvement in performance and cost compared to its predecessor. The release features stronger benchmark results, faster task completion, and shorter outputs, and is now live on the DeepSeek website, app, and API.
Enhanced Coding and Project Development
The model uses reinforcement learning with self-feedback to improve end-to-end project development, enabling users to build complete projects from scratch. Demonstrated capabilities include 3D modeling, web-based game creation, and complex scene generation, such as a Portal 2 recreation using the Godot engine. DeepSeek recommends upgrading to DeepSeek Harness v0.2.1-rc.1 or later for the best experience, noting optimizations for Tencent CodeBuddy and WorkBuddy.
Search & Fetch API Availability
DeepSeek’s Search & Fetch APIs are now open to everyone, powered by an in-house search engine that handles hundreds of millions of daily requests. The pricing is set at $0.75 per 1,000 calls for each endpoint. While calls via server-side tools in DeepSeek Harness, Claude Code, and Codex are currently free for DeepSeek API users, billing for these integrations begins on Oct 31, 2026, at 16:00 UTC.
Open-Source Release
The model weights are available for download on Hugging Face and ModelScope, continuing DeepSeek’s strategy of open-sourcing its key models.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.