AITOPAI热报
ModelsSelected73°

DeepSeek V4.1 Flash Model Update

向阳乔木@vista8
Post· 2 sources
View the post
Why we picked it

Priced same as V4 Flash with multimodal support, offering better value for developers in Southeast Asia working with visual data.

DeepSeek V4.1 Flash entered internal testing on 2 Sep 2026. The new model features a different architecture and native multimodal support. It costs the same as V4 Flash and allows up to 20 concurrent requests per account. Users can test it by changing the model ID to deepseek-v4.1-flash-expires-on-0910.

Original · 向阳乔木

据说 DeepSeek 官方微信群通知，DeepSeek V4.1 Flash内测了。 新模型换了新的模型架构，原生支持多模态，计费与 V4 Flash 相同，单账号最多 20 并发。 把模型 ID 换成 deepseek-v4.1-flash-expires-on-0910 就能测。 第二个我以为是幻觉，说穿条纹西装，仔细看了下果然有条纹... 💬 4 🔄 0 ❤️ 10 👀 6367 📊 5 ⚡

  • 深度求索 DeepSeek09-10 06:10原文
#DeepSeek#V4.1 Flash#multimodal#ai-models

Keep reading

Recent stories with the same tag or section
ProductsUS vs ChinaSelected07:26
vLLM sets best DeepSeek-V4.1-Flash result on SemiAnalysis AgentX

Matching DeepInfer's performance while beating it on a third-party agentic benchmark suggests vLLM is now a credible production backend for high-throughput DeepSeek serving, not just a research engine. Engineers running agent-heavy inference stacks can use the reproducible blog recipe to benchmark their own deployments against the 954K tokens/s figure. DeepSeek's off-peak pricing makes the revenue-per-GPU-hour number a useful cost baseline for capacity planning.

#vLLM#DeepSeek#AgentX#SemiAnalysis
官方账号vLLM@vllm_projectSource
ModelsBuildersSelected02:02
vLLM Matches DeepInfer on DeepSeek-V4.1-Flash Decode

An open-source inference stack catching up with a frontier lab's in-house engine means teams can serve DeepSeek-V4.1-Flash without relying on the vendor's own tooling. Engineers running self-hosted deployments in Singapore or ASEAN data centres can benchmark vLLM at DEP32 as a direct alternative to DeepInfer's EP64DP64 setup, avoiding engine lock-in while keeping comparable throughput at large batch sizes.

#vLLM#DeepSeek#DeepInfer#inference
官方账号vLLM@vllm_projectSource