LLM：Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".
active 2023-08-15 → 2026-06-07 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 1028 days from 2023-08-15 to 2026-06-07. Pushes: 0 total, peak 0 in a day. Pull requests: 7 total, peak 2 in a day. Issues: 19 total, peak 2 in a day. Comments: 10 total, peak 2 in a day. Stars: 1,063 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 |
|---|---|---|---|---|
| MilesQLi | 4 | 0 | 0 | 2 |
| igormolybog | 4 | 0 | 2 | 2 |
| efrantar | 3 | 0 | 0 | 3 |
| chandraprvkvsh | 2 | 0 | 2 | 0 |
| westboy123 | 2 | 0 | 0 | 0 |
| JachinJiang | 2 | 0 | 0 | 1 |
| amitsrivastava78 | 1 | 0 | 1 | 0 |
| charliebarber | 1 | 0 | 1 | 0 |
| benharvie | 1 | 0 | 0 | 0 |
| xd1073321804 | 1 | 0 | 0 | 0 |
| CXiaorong | 1 | 0 | 0 | 0 |
| shengqihailuo1 | 1 | 0 | 0 | 0 |
| nku-ligl | 1 | 0 | 0 | 0 |
| foricee | 1 | 0 | 0 | 0 |
| zhanyil2 | 1 | 0 | 0 | 0 |
| sayal-k | 1 | 0 | 0 | 1 |
| mynotwo | 1 | 0 | 0 | 0 |
| baiSongL | 1 | 0 | 0 | 0 |
| GuoYi0 | 1 | 0 | 0 | 0 |
| ROUJINN | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue#65nku-ligl2026-03-19 08:19Question about sampling 256 samples for C4 dataset evaluation
- Issue#64benharvie2026-03-08 05:21Responsible disclosure policy
- Pull request#62amitsrivastava782025-07-04 08:28
- Pull request#61charliebarber2025-05-08 13:40
- Issue comment#59sayal-k2025-04-02 12:00Problem about value range of valenc in func get_c4()
- Issue#60shengqihailuo12025-01-16 09:00Sensitivity analysis of model parameters
- Issue#59880999812024-11-13 07:18Problem about value range of valenc in func get_c4()
- Pull request#58chandraprvkvsh2024-11-08 05:42
- Pull request#58chandraprvkvsh2024-11-08 05:41
- Issue#57mjyun012024-10-10 12:22Reproducing Table3.
- Issue comment#21ROUJINN2024-07-02 08:02Why no update to Hinv
- Issue#56xd10733218042024-05-23 03:24Test on CNN model containing group conv by GPTQ method
- Issue#55puja932024-05-19 11:10Reconstruct Quantized Model Layer in torch.
- Pull request#54igormolybog2024-03-27 01:46
- Pull request#54igormolybog2024-03-27 01:46
- Issue#53zhanyil22024-02-13 16:54GPTQ on BERT based
- Issue#52GuoYi02024-01-30 04:29AssertionError
- Issue comment#51efrantar2023-12-29 11:12Regarding the method for computing the Hessian matrix.
- Issue#51baiSongL2023-12-29 08:35Regarding the method for computing the Hessian matrix.
- Issue#50CXiaorong2023-12-21 03:25GPTQ pseudo-quantization saved weights (pt format) How load Re-evaluation
- Issue#48mynotwo2023-12-10 03:28Compatibility of Quant3Linear and 4-bit quantization
- Issue#47frankxyy2023-12-03 04:33act-order on inference
- Issue#46westboy1232023-11-30 12:13pack_model takes too long time
- Issue#46westboy1232023-11-30 11:27pack_model takes too long time
- Issue comment#44MilesQLi2023-11-15 20:57H_inv not updated
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 1,063 stars here means stars gained during the window, not the repo's star count.