Open weights LLM from Meta
Python · active 2024-02-21 → 2026-08-04 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 896 days from 2024-02-21 to 2026-08-04. Pushes: 683 total, peak 23 in a day. Pull requests: 497 total, peak 13 in a day. Issues: 161 total, peak 10 in a day. Comments: 469 total, peak 15 in a day. Stars: 3,803 total, peak 417 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 |
|---|---|---|---|---|
| copybara-service[bot] | 998 | 650 | 346 | 0 |
| google-cla[bot] | 87 | 0 | 0 | 87 |
| Conchylicultor | 67 | 14 | 14 | 33 |
| Gopi-Uppari | 47 | 0 | 0 | 47 |
| tilakrayal | 27 | 0 | 0 | 14 |
| Balakrishna-Chennamsetti | 27 | 0 | 0 | 27 |
| hemanth | 13 | 0 | 0 | 7 |
| mehulbafnaa | 13 | 0 | 12 | 1 |
| FredrikNoren | 12 | 0 | 0 | 8 |
| Molugan | 11 | 4 | 2 | 3 |
| shreed27 | 8 | 0 | 4 | 2 |
| gustheman | 8 | 0 | 0 | 8 |
| carlofisicaro | 7 | 0 | 4 | 3 |
| pratik26x | 7 | 0 | 4 | 3 |
| markknoffler | 7 | 0 | 5 | 0 |
| prasad-yashdeep | 6 | 0 | 0 | 4 |
| syrkis | 6 | 0 | 1 | 2 |
| Vruddhi18 | 6 | 0 | 2 | 4 |
| DwarKapex | 5 | 0 | 0 | 2 |
| arJ-V | 5 | 0 | 3 | 1 |
Recent activity
Latest issues, pull requests and releases
- Pull request#691copybara-service[bot]2026-06-11 11:29
- Issue comment#671Ayushdevo2026-06-06 09:37ValueError: Cannot use apply_chat_template because this processor does not have a chat template.
- Pull request#667copybara-service[bot]2026-05-29 00:52
- Issue comment#659prince-shakyaa2026-05-23 23:39fix: np.float128 platform crash on macOS/Windows & missing f-string in error message
- Pull request#661prince-shakyaa2026-05-23 23:14
- Issue comment#659google-cla[bot]2026-05-23 22:41fix: np.float128 platform crash on macOS/Windows & missing f-string in error message
- Pull request#659prince-shakyaa2026-05-23 22:41
- Issue#658prince-shakyaa2026-05-23 22:36Bug: Missing f-prefix in error message in _normalize_token() makes stop_tokens debugging impossible
- Pull request#650copybara-service[bot]2026-05-20 13:33
- Issue comment#622imneme2026-05-19 01:33[BUG] Gemma 4 token repetition collapse during long generation — affects both 31b Dense and 26b MoE
- Pull request#649dparikh792026-05-16 23:43
- Pull request#648karlb-dev2026-05-16 02:42
- Pull request#645Rhabibi16092026-05-11 23:49
- Issue comment#622garciaautomation2026-05-06 05:49[BUG] Gemma 4 token repetition collapse during long generation — affects both 31b Dense and 26b MoE
- Issue comment#642tomohiro862026-05-04 00:37fix: use integer division (//) in GQA reshape operations
- Pull request#642tomohiro862026-05-04 00:36
- Pull request#642tomohiro862026-05-04 00:25
- Issue#641tomohiro862026-05-04 00:20Bug: float division instead of integer division in GQA reshape causes silent shape truncation
- Issue comment#632AlexandreManai2026-04-26 17:46Fix typo in LoRA fine-tuning comment
- Issue#628QuantumSorcerer022026-04-19 07:34Error fixes
- Issue#626rappstar2026-04-18 05:07[BUG] Gemma4:26B Applies Formatting/Linting Changes When Editing A Python File
- Issue comment#604blue-az2026-04-16 03:59Gemma 4 26B: strong protocol-following regression vs Gemma 3 in bounded machine-facing evaluation
- Issue comment#606rastc2026-04-11 08:43Error with `model = gm.nn.Gemma4_E2B()`
- Pull request#616copybara-service[bot]2026-04-07 09:41
- Issue comment#610Willis08262026-04-06 23:44[BUG] Gemma-4-26B-A4B: Deterministic Repetition Loop at 14th item in list ("Wait, I found it. The 14.")
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 3,803 stars here means stars gained during the window, not the repo's star count.