DeepSeek ships open-weights V4.1-Flash-1017 and opens its Search & Fetch APIs to all developers

DeepSeek released DeepSeek-V4.1-Flash-1017, an update to its V4.1-Flash model that is live on its website, app and API, with MIT-licensed weights on Hugging Face and ModelScope. It claims stronger benchmarks, faster task completion and shorter outputs, including 77.0 on DeepSWE v1.1 (up from 74.2). It also opened Search and Fetch APIs, powered by its in-house search engine, to everyone at $0.75 per 1,000 calls each.