Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level
TypeScript · active 2025-02-01 → 2026-08-08 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 554 days from 2025-02-01 to 2026-08-08. Pushes: 152 total, peak 13 in a day. Pull requests: 69 total, peak 5 in a day. Issues: 71 total, peak 5 in a day. Comments: 131 total, peak 7 in a day. Stars: 336 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 |
|---|---|---|---|---|
| giladgd | 307 | 152 | 62 | 61 |
| github-actions[bot] | 34 | 0 | 1 | 31 |
| onzag | 7 | 0 | 0 | 3 |
| albertkawmi | 5 | 0 | 2 | 1 |
| ido-pluto | 5 | 0 | 0 | 0 |
| louistiti | 4 | 0 | 0 | 4 |
| rogerdcarvalho | 4 | 0 | 0 | 4 |
| jakzo | 3 | 0 | 0 | 0 |
| paulthedev | 3 | 0 | 0 | 0 |
| goog | 3 | 0 | 0 | 0 |
| xinxi053 | 3 | 0 | 0 | 3 |
| snowyu | 3 | 0 | 0 | 1 |
| tx0h | 2 | 0 | 0 | 0 |
| makejiang | 2 | 0 | 0 | 1 |
| Fi1osof | 2 | 0 | 0 | 0 |
| briancullinan2 | 2 | 0 | 0 | 2 |
| andreinknv | 2 | 0 | 2 | 0 |
| ketansp | 2 | 0 | 0 | 1 |
| Griffork | 2 | 0 | 0 | 0 |
| raftario | 2 | 0 | 0 | 2 |
Recent activity
Latest issues, pull requests and releases
- Pull request#597CreatiCoding2026-05-24 00:33
- Pull request#606andreinknv2026-05-14 22:48
- Pull request#605andreinknv2026-05-14 22:46
- Issue#599Zighy2026-04-25 19:30bug: loadBackends(backendsPath) skipped when buildGpu === false, silently drops custom GPU backends compiled via NODE_LLAMA_CPP_CMAKE_OPTION_*
- Issue#594giladgd2026-04-12 20:27feat: `gemma4` model architecture
- Issue#593yurivict2026-04-08 18:37bug: Compilation fails on 32-bit platforms: error: use of overloaded operator '[]' is ambiguous (with operand types 'Napi::Uint32Array' (aka 'TypedArrayOf<unsigned int>') and 'int')
- Pull request#591giladgd2026-04-06 20:17
- Pull request#591giladgd2026-04-06 20:17
- Issue#584giladgd2026-04-05 16:45feat: Support native logprobs in `LlamaChatSession.prompt()` to avoid double inference
- Issue comment#554FringeNet2026-03-24 17:06Segmentation fault on CPU fallback when Vulkan prebuilt binary is incompatible (Linux x64, no GPU)
- Pull request#582giladgd2026-03-17 06:36
- Issue#583giladgd2026-03-17 04:57feat: Support for Paged KV Cache and Hardware-Native KV Quantization (NVFP4/Blackwell)
- Issue#583tx0h2026-03-16 21:56feat: Support for Paged KV Cache and Hardware-Native KV Quantization (NVFP4/Blackwell)
- Issue#583tx0h2026-03-16 21:56feat: Support for Paged KV Cache and Hardware-Native KV Quantization (NVFP4/Blackwell)
- Issue#575github-actions[bot]2026-03-15 21:18bug: invalid output when using grammar
- Issue comment#570github-actions[bot]2026-03-15 21:18fix(`resolveModelFile`): deduplicate concurrent downloads
- Pull request#570github-actions[bot]2026-03-15 21:18
- Issue#580enceos2026-03-15 08:27Windows CUDA: cuMemAddressReserve failure in VMM pool causes hard abort (GGML_CUDA_NO_VMM workaround)
- Issue#571giladgd2026-03-15 01:00bug: Qwen3.5 Cannot Run with GPU Acceleration on Apple M4
- Issue#575giladgd2026-03-15 01:00bug: invalid output when using grammar
- Issue#577giladgd2026-03-15 01:00bug: version >=v3.15.0 not support CUDA v13.1 & v13.2
- Issue comment#577giladgd2026-03-15 00:45bug: version >=v3.15.0 not support CUDA v13.1 & v13.2
- Issue comment#577xinxi0532026-03-15 00:32bug: version >=v3.15.0 not support CUDA v13.1 & v13.2
- Issue comment#577giladgd2026-03-14 15:46bug: version >=v3.15.0 not support CUDA v13.1 & v13.2
- Issue comment#577xinxi0532026-03-14 00:48bug: version >=v3.15.0 not support CUDA v13.1 & v13.2
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 336 stars here means stars gained during the window, not the repo's star count.