AITOPAI热报
ProductsSelected

vLLM 在 AgentX 基准创下 DeepSeek-V4.1-Flash 最佳成绩

官方账号vLLM@vllm_project
Post
View the post
Why we picked it

vLLM 又刷纪录了：跑 DeepSeek-V4.1-Flash 智能体负载，单卡每小时能赚 38.96 美元，博客还教你怎么复现。

vLLM 团队在 SemiAnalysis 的 AgentX 基准上测试了真实智能体工作负载，跑出 DeepSeek-V4.1-Flash 的历史最佳结果。在 DeepSeek 非高峰定价下，单 GPU 每小时收入达 38.96 美元，总吞吐量为每 GPU 954K tokens/s（含缓存、输入与输出），同时单用户 P90 保持在 196 tokens/s。团队还表示性能已追平 DeepInfer，优化细节和复现方法已发布在官方博客。

Original · vLLM

Beyond matching DeepInfer's performance, we also put vLLM to the test on @SemiAnalysis_ AgentX with real-world agentic workloads.

At DeepSeek's off-peak pricing, vLLM delivers $38.96/GPU-hour revenue, with 954K tokens/s/GPU total throughput (cached+input+output) at 196 tokens/s P90 per user.

This is the best DeepSeek-V4.1-Flash result on AgentX ever, powered by recent advancement in vLLM!

Check out our blog for the full optimization story and how to reproduce the results 👇 Enjoy the read.

#vLLM#DeepSeek#AgentX#推理优化#开源模型

Keep reading

Recent stories with the same tag or section
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
ModelsBuildersSelected11:25
MiniMax M3.1 Flash Runs in Production on vLLM

Production-grade serving of a MiniMax model through vLLM signals that the open-source inference stack can now carry commercial workloads with SLA guarantees. Teams evaluating cheaper open-weight serving options can look at Inferact's vLLM-based offering as a reference setup. Upstream contributions also mean the wider vLLM community inherits the reliability hardening that came out of this deployment.

#MiniMax#vLLM#open-source models#inference
官方账号vLLM@vllm_projectSource