Deepseek releases V4.1-Flash-1017, posts open weights and opens search APIs
Deepseek released its V4.1-Flash-1017 model and published open weights on Hugging Face and ModelScope, claiming faster, shorter outputs and stronger coding and agent-task scores. Internal benchmarks show the small Flash model beating larger rivals while cutting per-task cost to about $0.10 to $0.20. Deepseek also opened its in-house Search and Fetch APIs to outside developers.
Where this came from. 2 outlets reported this. We read them, then wrote the sixty words above.