Minimal GPU runtime for Python - high-performance CUDA kernels, memory management, and LLM inference without heavy dependencies
active 2025-12-11 → 2026-02-11 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 63 days from 2025-12-11 to 2026-02-11. Pushes: 148 total, peak 17 in a day. Pull requests: 30 total, peak 10 in a day. Issues: 120 total, peak 31 in a day. Comments: 43 total, peak 12 in a day. Stars: 0 total, peak 0 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 |
|---|---|---|---|---|
| m96-chan | 341 | 148 | 30 | 43 |
Recent activity
Latest issues, pull requests and releases
- Pull request#194m96-chan2026-02-11 16:41
- Releasem96-chan2026-01-01 19:25v0.2.19 - FLUX.1 Image Generation
- Issue#187m96-chan2026-01-01 19:14perf(diffusion): FLUX.1 transformer performance optimization
- Issue#177m96-chan2026-01-01 19:08RFC: Image Generation Support (Stable Diffusion, Flux, DiT)
- Pull request#185m96-chan2026-01-01 13:07
- Issue#184m96-chan2026-01-01 12:35test(tts): Verify Kokoro TTS implementation (#183)
- Issue#179m96-chan2026-01-01 12:35bug(tts): Kokoro TTS outputs 440Hz sine wave instead of speech
- Pull request#183m96-chan2026-01-01 12:27
- Issue#181m96-chan2026-01-01 11:43perf(llm): Batch decode missing CUDA Graph and zero-alloc optimizations
- Issue#181m96-chan2026-01-01 11:43perf(llm): Batch decode missing CUDA Graph and zero-alloc optimizations
- Issue#181m96-chan2026-01-01 11:43perf(llm): Batch decode missing CUDA Graph and zero-alloc optimizations
- Issue#150m96-chan2025-12-31 03:14feat(profiling): GPU kernel profiler and memory analyzer
- Issue comment#159m96-chan2025-12-31 03:14feat(core): Memory-Mapped Model Loading
- Issue comment#150m96-chan2025-12-31 03:14feat(profiling): GPU kernel profiler and memory analyzer
- Issue comment#159m96-chan2025-12-31 02:18feat(core): Memory-Mapped Model Loading
- Pull request#175m96-chan2025-12-31 01:45
- Issue#118m96-chan2025-12-30 23:29[v0.2.17]feat: FP8 model loading and inference support
- Issue comment#118m96-chan2025-12-30 23:29[v0.2.17]feat: FP8 model loading and inference support
- Releasem96-chan2025-12-30 22:59v0.2.18
- Issue#171m96-chan2025-12-30 22:51feat(nn): Add ReLU² (Squared ReLU) activation function
- Issue comment#171m96-chan2025-12-30 22:51feat(nn): Add ReLU² (Squared ReLU) activation function
- Issue#167m96-chan2025-12-30 22:51feat(nn): Add ALiBi (Attention with Linear Biases) positional encoding
- Issue comment#167m96-chan2025-12-30 22:51feat(nn): Add ALiBi (Attention with Linear Biases) positional encoding
- Pull request#174m96-chan2025-12-30 22:49
- Pull request#174m96-chan2025-12-30 22:25
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 0 stars here means stars gained during the window, not the repo's star count.