Model card comparison: DeepSWE 77.0 beats Opus 5's 74.0

Qwen AI Fans @QwenAIFansOriginally posted 2026.10.17

Open original on X

DeepSeek's "minor update" to V4.1 Flash is open weights, MIT, same day as the API. from their own model card: DeepSWE: 77.0 vs Claude Opus 5's 74.0 Agents' Last Exam: 35.4 vs 28.6 Terminal-Bench 4.0: 42.3 vs 51.8 HLE: 42.6 vs 56.3 ahead of Opus on agent work, still behind on hard reasoning. and the outputs got shorter, so it's cheaper per task too source: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017 #DeepSeek #OpenSourceAI