← Back to live feed1 story
DeepSeek's newest model reached another platform later the same day, with Venice saying users can access V4.1 Flash there and that the service is fully private.
DeepSeek introduced V4.1 Flash on Sept. 10 and said the model has native multimodal support and is live on its API. The company said all V4 Pro API requests will route to V4.1 Flash starting Sept. 14 until V4.1 Pro launches.
Sign in to suggest edits
Key sources
- SOURCE@askvenice“fully private.”x.com
- SOURCE@justinsuntron“Supporting a 1M context window and up to 384K output”x.com
- SUPPORT@wallstengine“Its global KV cache is down to 890 bytes per token, versus 3,514 for V4-Flash and 48,068 for V3.2.”x.com
- SUPPORT@therundownai“It beats GPT-5.6 Sol and Claude Opus 5 on several agentic, coding, and cyber benchmarks.”x.com
- SUPPORT@arena“Among open models, DeepSeek-V4.1-Flash landed at ~#4 within 11 pts of Qwen3.8-Flash-Next.”x.com
- SUPPORT@valsai“It beats Kimi K3 while costing $0.30 a test, the cheapest model in the open-weight top 10.”x.com
- SUPPORT@yuchenj_uw“Databricks will be bringing it to our customers soon!”x.com
- SUPPORT@cline“it's $0.30/$1.20 per 1M input/output.”x.com