Home / Open source / vllm
vllm
vllm-project/vllm · Enterprise stack · Enterprise
A high-throughput and memory-efficient inference and serving engine for LLMs
Verified via GitHub API
- gpt
- llm
- pytorch
- model-serving
- transformer
- llm-serving
- inference
- llama
- amd
- cuda
- tpu
- deepseek
- qwen
- blackwell
- deepseek-v3
- gpt-oss
- kimi
- moe
- openai
- qwen3
Data from GitHub API, refreshed Oct 17, 2026
- Stars
- 93969
- Forks
- 23480
- License
- Apache-2.0
- Language
- Python
- Last commit
- Oct 17, 2026
- Latest release
- v0.31.0
- Release date
- Oct 5, 2026
- Contributors
- 3780
- Open issues
- 2647
Works with — from README
- Models
- DeepSeek, Llama, Mistral, OpenAI, Qwen
Source: README.md keyword match (stated in README)
vllm FAQ
Is vllm free?
vllm is open source under the Apache-2.0 license. Hosted services and infrastructure may have separate costs.
How popular is vllm?
vllm has 93969 GitHub stars as of Oct 17, 2026.
Is there a hosted version of vllm?
A hosted alternative for vllm is unknown (—) in this data.
What does vllm work with?
From README: DeepSeek, Llama, Mistral, OpenAI, Qwen.