zhihu / ZhiLight
A highly optimized LLM inference acceleration engine for Llama and its variants.
Why this job: says: llm inference; topics: llm-inference, inference-engine, llm-serving, model-serving, llama
What it does
- Asynchronous OpenAI compatible interface adapted from vllm
- Custom defined tensor and unified global memory management
- Encode and all-reduce overlap, we named "dual streams"
- Host all-reduce based on SIMD instructions
- Optimized fused kernels, qkv, residual & layernorm etc.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Inactive
- Last commit 7 months ago
- Stars
- 906
- On GitHub
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.4.8
- Released 23 months ago
- Built with
- C++
- 67% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 71% closed
- 5 still open
- Pull requests
- 62 merged
- 1 waiting
- Releases
- 1
- 0 in the last year
- No contributing guide
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| vllm-project / vllm | 96,774 | today | Apache-2.0 | Mac, Linux |
| alibaba / rtp-llm | 1,394 | yesterday | Apache-2.0 | No downloads |
| liguodongiot / llm-action | 25,274 | 29 days ago | Apache-2.0 | No downloads |
More: languages, topics and project record
Languages
- C++ 67%
- Cuda 22%
- Python 10%
- C 1%
Topics
- cuda
- deepseek-r1
- gpt
- inference-engine
- llama
- llm
- llm-inference
- llm-serving
- model-serving
- pytorch
Project record
- JobLocal LLMs
- StatusListed
- Commits111
- Forks105
- Watching52
- Stars in 28 days0
- CreatedDec 6, 2024
- First seen by usNov 7, 2026
- CheckedNov 8, 2026