[ICLR2024 spotlight] OmniQuant is a simple and powerful quantization technique for LLMs.
active 2024-07-29 → 2026-06-07 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 679 days from 2024-07-29 to 2026-06-07. Pushes: 1 total, peak 1 in a day. Pull requests: 1 total, peak 1 in a day. Issues: 14 total, peak 2 in a day. Comments: 24 total, peak 3 in a day. Stars: 156 total, peak 8 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| ChenMnZ | 6 | 1 | 0 | 4 |
| forcekkk | 3 | 0 | 0 | 2 |
| mostafaelhoushi | 2 | 0 | 0 | 2 |
| Wotoosh | 2 | 0 | 0 | 2 |
| pavelgolikov | 2 | 0 | 0 | 0 |
| zjq0455 | 2 | 0 | 0 | 0 |
| stackByStack | 2 | 0 | 0 | 1 |
| WX-yh | 2 | 0 | 0 | 0 |
| jiuyixu25 | 2 | 0 | 0 | 1 |
| threeEggPie | 1 | 0 | 0 | 1 |
| FlexLaughing | 1 | 0 | 0 | 0 |
| TokuyuSou | 1 | 0 | 1 | 0 |
| Hzqskywkr | 1 | 0 | 0 | 0 |
| kimoji919 | 1 | 0 | 0 | 1 |
| JustVelkhana | 1 | 0 | 0 | 0 |
| SSshuishui | 1 | 0 | 0 | 1 |
| ZijianYY | 1 | 0 | 0 | 1 |
| trinks-slam818 | 1 | 0 | 0 | 1 |
| Tfloow | 1 | 0 | 0 | 1 |
| shubhra | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#114Tfloow2026-04-11 11:48update: support new Llama API + assess OpenGVLab/OmniQuant#113
- Issue comment#110jiuyixu252026-01-16 01:07ModuleNotFoundError: No module named 'auto_gptq'
- Issue#110jiuyixu252026-01-16 00:28ModuleNotFoundError: No module named 'auto_gptq'
- Issue#109yz3022025-12-16 02:48Can omniquant be used on CNN models, similar to YOLO?
- Pull request#108TokuyuSou2025-09-17 07:40
- Issue comment#97B0B8K1ng2025-08-14 11:59Add support for Llama3.1
- Issue#106JustVelkhana2025-06-21 05:41A single A100-80G can't run Llama-2-70b model?
- Issue comment#70threeEggPie2025-04-17 06:28W4A4 in llama2-7b
- Issue#105forcekkk2025-04-11 05:30RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:7 and cuda:0! (when checking argument for argument index in method wrapper_CUDA__index_select)
- Issue comment#89forcekkk2025-04-11 04:27RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat2 in method wrapper_CUDA_bmm)
- Issue comment#89forcekkk2025-04-11 04:17RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat2 in method wrapper_CUDA_bmm)
- Issue comment#93luo13zhi2025-02-22 09:15The llama-2-7b model can't quant in this code
- Issue comment#104mostafaelhoushi2025-01-26 05:13I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b. It happened in int_llama_layer.py: self.rotary_emb = copy.deepcopy(org_module.rotary_emb)
- Issue comment#103mostafaelhoushi2025-01-26 05:13I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b.
- Issue comment#104trinks-slam8182025-01-05 05:04I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b. It happened in int_llama_layer.py: self.rotary_emb = copy.deepcopy(org_module.rotary_emb)
- Issue#104WX-yh2025-01-05 04:58I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b. It happened in int_llama_layer.py: self.rotary_emb = copy.deepcopy(org_module.rotary_emb)
- Issue#103WX-yh2025-01-05 04:54I encounter a error: "AttributeError: 'LlamaAttention' object has no attribute 'rotary_emb'",when i run code with llama-1-7b.
- Issue comment#102stackByStack2024-12-16 04:06Fail to reproduce the result of w2a16 using llama2 7b
- Issue#102stackByStack2024-12-16 04:02Fail to reproduce the result of w2a16 using llama2 7b
- Issue comment#101ChenMnZ2024-11-21 00:33Will the qwen2.5 model be supported in the future？
- Issue comment#100ChenMnZ2024-11-21 00:30OmniQuant belong to PTQ or QAT？
- Issue#94ChenMnZ2024-11-01 01:34Performance gap with Llama-2-7B
- Issue comment#81SherrySwift2024-10-31 06:22Questions about quantization
- Issue comment#8Wotoosh2024-10-12 15:22why quantize opt-1.3b or llama 7b with W8A8 config loss is nan?
- Issue comment#8ChenMnZ2024-10-12 14:27why quantize opt-1.3b or llama 7b with W8A8 config loss is nan?
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 156 stars here means stars gained during the window, not the repo's star count.