DeepSeek Releases V4.1-Flash-1017, Claiming Improved Efficiency and Benchmark Gains
DeepSeek says the model is live on its website, app and API, and optimized for DeepSeekHarness and Tencent CodeBuddy.
TLDR
DeepSeek says V4.1-Flash-1017 delivers stronger benchmark performance, faster task completion and shorter outputs than the previous version. It recommends upgrading to DeepSeek Harness v0.2.1-rc.1 or later for the best experience. A separate post claims gains across agent and coding benchmarks without a price increase, including a Terminal-Bench 4.0 score rising from 31.2 to 42.3.
Combined views
421K
24 Sources, first seen ago
