vllm-ascend
vllm-ascend is What this PR does / why we need it? Add the Hy4 W8A8-MXFP8 deployment guide for Ascend 950DT, covering single-node, multi-node and Prefill-Decode deployments. Update the support matrix and document.... It is ranked #240 on the AI Radar, in AI infra & evals, first seen 3 days ago and shared in 3 posts (32.3k views).
[Doc] Add Hy4 deployment guide by ZYY-ing · Pull Request #20671 · vllm-project/vllm-ascend · GitHub Skip to content Navigation Menu Sign in Search / Sign in Sign up Appearance settings You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Repository navigation [Doc] Add Hy4 deployment guide - # 20671 # 20671 Open ZYY-ing wants to merge 13 commits into vllm-project:main vllm-project/vllm-ascend:main from ZYY-ing:main ZYY-ing/vllm-ascend:main…
What people said about vllm-ascend on X
🚨重磅！腾讯混元Hy4正式版即将发布，真实生产力将全面升级！官方已开启预热！ 关键看点： 🔹基于Hy4 Preview全面升级，正式版即将登场 🔹首次加入视觉理解（To see👀），支持看图分析 🔹强化长程任务、Agent能力与思维链效率 🔹聚焦代码/办公/科研等真实生产力场景 🔹Hy4 Flash同步预告，主打高性价比 To see, To create, To work, To research。 #腾讯混元 #Hy4 #混元大模型 #AI生产力 #视觉理解 #WorkBuddy #姚顺雨 #马化腾…
— @NFT_Chen, 3 days ago · 91 likes · see the post
🚨重磅预警！腾讯混元Hy4正式版即将上线！ 视觉能力首次上车，部署信号已亮起！ 拆解 @aitrackerbot 探测关键变动如下： 🔹参数规模不变：仍是770B总参数 / 49B激活的MoE架构（与Hy4 preview一致），另带10B原生MTP层（支持投机解码加速） 🔹最大新增：首次引入视觉理解能力（image input），正式补上多模态短板 🔹能力强化：重点提升长程任务、真实生产力场景（编程/办公/科研）表现，并优化思维链效率 🔹轻量版本同步预告：Hy4 flash即将推出，继承视觉+Agent能力，走极致性价比路线…
— @NFT_Chen, 3 days ago · 54 likes · see the post
Tencent Hy4 is out in Tencent Cloud's official API 🐧 The final version follows the August preview: 1M context, 64K output, and vision for the first time. On Vercel AI Gateway it's $0.42 / $1.25 per 1M tokens, half the preview's price. Weights aren't public yet.
— @aitrackerbot, 2 days ago · 56 likes · see the post
Alternatives to vllm-ascend
- Spark-X2.6 — 6 open routing models from 0.3B to 27B, built to make decisions and extract exactly what you need in a single call. 🤖
- fireworks.ai — Start building through our API, no dedicated GPUs needed
- OrcaRouter — Claim yours
- aimlapi.com — Crawl Reddit with your AI agent >
- LongCatAI — Take a closer look at what it can do
- Kimi API Platform — Kimi API Platform, providing the 2.8-trillion-parameter Kimi K3 large language model API with a 1M-token context…
vllm-ascend in numbers
- Rank on the AI Radar: #240 of 10412
- Shared in 3 posts by 2 accounts: @NFT_Chen, @aitrackerbot
- 32.3k views on those posts
- First seen 3 days ago, last shared 2 days ago
- Pricing seen by AGK Intelligence: open source
- Market: AI infra & evals
FAQ
What is vllm-ascend?
What this PR does / why we need it? Add the Hy4 W8A8-MXFP8 deployment guide for Ascend 950DT, covering single-node, multi-node and Prefill-Decode deployments. Update the support matrix and document... It was first shared on X 3 days ago and is ranked #240 on the AI Radar.
Is vllm-ascend free?
It is open source.
Who shared vllm-ascend?
2 accounts on X, including @NFT_Chen, @aitrackerbot, in 3 posts totalling 32.3k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.5k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 16:50 UTC. Full method.