Automatically split your PyTorch models on multiple GPUs for training & inference
active 2023-08-16 → 2026-01-22 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 891 days from 2023-08-16 to 2026-01-22. Pushes: 0 total, peak 0 in a day. Pull requests: 2 total, peak 1 in a day. Issues: 21 total, peak 2 in a day. Comments: 21 total, peak 4 in a day. Stars: 315 total, peak 7 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 |
|---|---|---|---|---|
| Ar-Kareem | 6 | 0 | 1 | 4 |
| JunyiYe | 3 | 0 | 0 | 1 |
| BlackSamorez | 3 | 0 | 0 | 2 |
| Dr-Left | 2 | 0 | 0 | 1 |
| longday1102 | 2 | 0 | 0 | 2 |
| yxk9810 | 2 | 0 | 0 | 0 |
| yyya9 | 2 | 0 | 0 | 1 |
| ReinForce-II | 2 | 0 | 1 | 0 |
| MonolithFoundation | 1 | 0 | 0 | 0 |
| PieterZanders | 1 | 0 | 0 | 1 |
| ludwigflo | 1 | 0 | 0 | 0 |
| 342215448 | 1 | 0 | 0 | 0 |
| HanGyeol-Yoo | 1 | 0 | 0 | 0 |
| smallmocha | 1 | 0 | 0 | 0 |
| zhangjunyi111 | 1 | 0 | 0 | 1 |
| deema-A | 1 | 0 | 0 | 1 |
| jameswu2014 | 1 | 0 | 0 | 0 |
| chujiezheng | 1 | 0 | 0 | 1 |
| dutsc | 1 | 0 | 0 | 1 |
| GeneZC | 1 | 0 | 0 | 1 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#84Tezcan982024-10-08 14:10Does tensor_parallel support multi-node tensor parallel training?
- Issue comment#137Dr-Left2024-10-08 04:16Compatibility with `transformers > 4.36`: error: `AttributeError: 'tuple' object has no attribute 'to_legacy_cache'`
- Issue comment#137lybbill2024-10-02 10:18Compatibility with `transformers > 4.36`: error: `AttributeError: 'tuple' object has no attribute 'to_legacy_cache'`
- Issue#137Dr-Left2024-08-21 09:52Compatibility with `transformers > 4.36`: error: `AttributeError: 'tuple' object has no attribute 'to_legacy_cache'`
- Issue comment#84deema-A2024-08-14 03:25Does tensor_parallel support multi-node tensor parallel training?
- Issue comment#84PieterZanders2024-05-10 19:47Does tensor_parallel support multi-node tensor parallel training?
- Issue#136MonolithFoundation2024-05-01 10:54Customized generate func support?
- Issue comment#121GeneZC2024-04-19 03:54RuntimeError: NCCL Error 3: internal error
- Issue#135Qiovo0662024-03-22 06:16when I use transforers==3.7.0，ValueError: TensorParallelPreTrainedModel does not support Flash Attention 2.0 yet。
- Issue comment#66sgsdxzy2024-03-15 04:39Slow inference performance for large Llama models compared to naive MP
- Issue comment#66dutsc2024-03-15 02:55Slow inference performance for large Llama models compared to naive MP
- Pull request#134ReinForce-II2024-01-02 10:23
- Issue#133ReinForce-II2024-01-02 08:43tensor_parallel int4 LLM is not working since release v2.0.0
- Issue#132HanGyeol-Yoo2023-12-31 09:33Now, does tensor_parallel no longer support the huggingface trainer?
- Issue comment#66chujiezheng2023-11-26 00:10Slow inference performance for large Llama models compared to naive MP
- Issue#1303422154482023-11-14 12:08No implement of generate() when using models from
- Issue#129toufunao2023-11-03 07:02TensorParallel object has no attribute save_pretrained
- Issue comment#128yyya92023-10-27 13:51No output when using tensor_parallel
- Issue#128yyya92023-10-27 13:49No output when using tensor_parallel
- Issue#127yxk98102023-10-19 08:38How to use the model in a scenario where it is stored in the Safetenors format?
- Issue#127yxk98102023-10-19 02:47How to use the model in a scenario where it is stored in the Safetenors format?
- Issue#126JunyiYe2023-10-14 19:44Out of GPU memory for two A10 GPUs
- Issue comment#126JunyiYe2023-10-14 19:44Out of GPU memory for two A10 GPUs
- Issue#126JunyiYe2023-10-11 22:07Out of GPU memory for two A10 GPUs
- Issue#125QiueY5142023-10-09 01:13AttributeError: object has no attribute 'devices'
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 315 stars here means stars gained during the window, not the repo's star count.