（能在笔记本上运行 LLaMA 大模型）Port of Facebook's LLaMA model in C/C++
active 2023-08-15 → 2025-02-15 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 551 days from 2023-08-15 to 2025-02-15. Pushes: 3,808 total, peak 59 in a day. Pull requests: 4,145 total, peak 40 in a day. Issues: 4,715 total, peak 210 in a day. Comments: 26,022 total, peak 254 in a day. Stars: 24,770 total, peak 176 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 |
|---|---|---|---|---|
| ggerganov | 6.3K | 1.8K | 1.1K | 2K |
| github-actions[bot] | 4.1K | 0 | 0 | 2.2K |
| slaren | 3.3K | 641 | 316 | 1.5K |
| ngxson | 1.3K | 76 | 154 | 684 |
| JohannesGaessler | 1.2K | 85 | 183 | 753 |
| compilade | 1K | 173 | 57 | 446 |
| cebtenzzre | 910 | 73 | 79 | 470 |
| KerfuffleV2 | 845 | 19 | 54 | 630 |
| phymbert | 804 | 231 | 58 | 342 |
| 0cc4m | 544 | 86 | 66 | 283 |
| Cebtenzzre | 507 | 22 | 65 | 270 |
| mofosyne | 416 | 48 | 72 | 197 |
| NeoZhangJianyu | 390 | 21 | 43 | 225 |
| staviq | 318 | 0 | 5 | 272 |
| klosax | 311 | 63 | 5 | 210 |
| Galunid | 305 | 13 | 32 | 160 |
| ochafik | 300 | 18 | 58 | 152 |
| teleprint-me | 295 | 0 | 14 | 233 |
| ikawrakow | 293 | 72 | 67 | 120 |
| airMeng | 280 | 51 | 40 | 112 |
Recent activity
Latest issues, pull requests and releases
- Pull request#11884708-1452025-02-15 12:40
- Issue comment#11016ochafik2025-02-15 11:49Add Jinja template support
- Issue comment#11873VanceVagell2025-02-15 11:41Feature Request: Use all (or configurable #) of threads for model loading, not constrainted by --threads specified for inference
- Releasegithub-actions[bot]2025-02-15 10:45b4721
- Issue comment#5763ggerganov2025-02-15 10:28llama : add T5 (encoder-decoder) support
- Pull request#11880ochafik2025-02-15 10:11
- Issue comment#11880ochafik2025-02-15 10:11tool-call: fix type promotion typo causing crashes w/ --jinja w/o tools
- Issue comment#5763fairydreaming2025-02-15 10:07llama : add T5 (encoder-decoder) support
- Issue comment#11769remyoudompheng2025-02-15 10:07vulkan: implement several ops relevant for ggml_opt
- Issue comment#11880ochafik2025-02-15 10:05tool-call: fix type promotion typo causing crashes w/ --jinja w/o tools
- Issue comment#11016ggerganov2025-02-15 09:26Add Jinja template support
- Issue comment#5763JohannesGaessler2025-02-15 09:05llama : add T5 (encoder-decoder) support
- Issue comment#11873ggerganov2025-02-15 08:35Feature Request: Use all (or configurable #) of threads for model loading, not constrainted by --threads specified for inference
- Issue#11083bobleer2025-02-15 08:35Misc. bug: [Mac M4]llama-server cannot run in release-4409 but can run in 4406
- Releasegithub-actions[bot]2025-02-15 08:34b4720
- Issue comment#118150cc4m2025-02-15 08:33Improved KV cache loading performance for Vulkan, resulting in a 20x …
- Issue comment#117690cc4m2025-02-15 08:30vulkan: implement several ops relevant for ggml_opt
- Issue comment#11880ggerganov2025-02-15 08:25tool-call: fix --jinja w/o tools (type promotion typo caused crashes)
- Issue comment#8616DonaldCMLIN2025-02-15 08:24Bug: When llama-server.exe or llama-cli.exe is executed without any error messages appearing, it may not be operating normally when GGUF models are enabled.
- Issue comment#11788foldl2025-02-15 08:20Compile bug: can't build with MSVC + Vulkan
- Issue comment#5763fairydreaming2025-02-15 08:12llama : add T5 (encoder-decoder) support
- Issue comment#11880MoonRide3032025-02-15 08:01tool-call: fix --jinja w/o tools (type promotion typo caused crashes)
- Pull request#115280cc4m2025-02-15 08:01
- Pull request#109890cc4m2025-02-15 07:41
- Issue comment#11866MoonRide3032025-02-15 07:23Misc. bug: Problems with official jinja templates (Gemma 2, Llama 3.2, Qwen 2.5)
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 24,770 stars here means stars gained during the window, not the repo's star count.