implement custom models
active 2023-08-15 → 2026-05-14 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 1004 days from 2023-08-15 to 2026-05-14. Pushes: 4,224 total, peak 953 in a day. Pull requests: 46 total, peak 6 in a day. Issues: 103 total, peak 5 in a day. Comments: 222 total, peak 9 in a day. Stars: 59 total, peak 2 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 |
|---|---|---|---|---|
| hannanjgaws | 2.4K | 2.4K | 7 | 16 |
| jluntamazon | 993 | 980 | 4 | 5 |
| aws-yishanm | 854 | 854 | 0 | 0 |
| dacorvo | 44 | 0 | 0 | 27 |
| awshaichen | 28 | 20 | 6 | 0 |
| aws-rhsoln | 21 | 0 | 0 | 14 |
| jeffhataws | 21 | 0 | 0 | 15 |
| liechtym | 17 | 0 | 0 | 10 |
| awsilya | 16 | 0 | 0 | 12 |
| aws-guangyhu | 13 | 0 | 0 | 13 |
| mrnikwaws | 13 | 0 | 0 | 8 |
| enochlev | 11 | 0 | 0 | 6 |
| aws-donkrets | 11 | 0 | 0 | 10 |
| gsnaws | 10 | 0 | 1 | 7 |
| aws-mesharma | 9 | 3 | 3 | 0 |
| kahfizulkifli | 9 | 0 | 2 | 4 |
| sssrijan-amazon | 8 | 0 | 3 | 1 |
| yogendra-yatnalkar | 6 | 0 | 0 | 2 |
| michaelfeil | 6 | 0 | 0 | 4 |
| dennj | 6 | 0 | 2 | 2 |
Recent activity
Latest issues, pull requests and releases
- Pull request#101sssrijan-amazon2025-10-02 23:20
- Pull request#93sssrijan-amazon2025-10-02 23:19
- Pull request#108sssrijan-amazon2025-10-02 22:09
- Issue comment#47aws-guangyhu2025-10-02 21:54Rebasing PR into main (Adding support and performance link)
- Issue comment#56aws-guangyhu2025-10-02 21:51Turn off safe_serialization from save_split so that save_function is called
- Issue comment#47aws-guangyhu2025-10-02 21:51Rebasing PR into main (Adding support and performance link)
- Issue comment#86aws-guangyhu2025-10-02 21:51For Mistral 7B - Generate Text using Input Embeddings + Add no_repeat_ngram_size Support
- Issue comment#87aws-guangyhu2025-10-02 21:51Add Gemma
- Issue comment#101aws-guangyhu2025-10-02 21:50Support for Qwen2 (Llama2 based)
- Issue comment#106aws-guangyhu2025-10-02 21:50B & S transpose issue in BSH collectives_layout MLP output
- Issue comment#108aws-guangyhu2025-10-02 21:50Incorrect all-to-all layout transformation implementation
- Issue comment#109aws-guangyhu2025-10-02 21:50Incorrect splitting of axes to retain sharding of tensors
- Issue#88sssrijan-amazon2025-10-01 19:52llava support
- Issue#72sssrijan-amazon2025-10-01 19:51Generate Llama 2 from Embeddings
- Issue#80sssrijan-amazon2025-10-01 19:50Improve Neuron model loading time
- Issue comment#76aws-guangyhu2025-10-01 19:06User feedback when compiling and reloading a large model
- Issue comment#73aws-guangyhu2025-10-01 19:06Infering logits from `model.forward` for the entire batch instead of the last forward's output.
- Issue comment#72aws-guangyhu2025-10-01 19:04Generate Llama 2 from Embeddings
- Issue comment#92aws-guangyhu2025-10-01 19:01Not able to load llama 3 70b on inf2.24xlarge instance
- Issue comment#84sssrijan-amazon2025-09-15 21:52Latest changes introduced for continuous batching break Mixtral model
- Issue#92sssrijan-amazon2025-09-15 21:48Not able to load llama 3 70b on inf2.24xlarge instance
- Issue comment#94karthickgopalswamy2025-09-05 22:07Gibberish output for princeton-nlp/Sheared-LLaMA-1.3B with continuous batching
- Issue#94karthickgopalswamy2025-09-05 22:07Gibberish output for princeton-nlp/Sheared-LLaMA-1.3B with continuous batching
- Issue#78micwade-aws2025-09-04 21:04Backward compatibility with saved llama 2 compiled artifacts
- Issue comment#28micwade-aws2025-09-04 20:19Discrepancies Between GPU and Neuron-based Outputs for GPTJ Model on inf2.24xlarge
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 59 stars here means stars gained during the window, not the repo's star count.