active 2025-11-01 → 2026-04-03 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 154 days from 2025-11-01 to 2026-04-03. Pushes: 25 total, peak 4 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 78 total, peak 16 in a day. Comments: 25 total, peak 9 in a day. Stars: 3 total, peak 1 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 |
|---|---|---|---|---|
| akaszubski | 128 | 25 | 0 | 25 |
Recent activity
Latest issues, pull requests and releases
- Issue#86akaszubski2026-04-03 03:06refactor: remove redundant truncation fallback
- Issue comment#86akaszubski2026-04-03 03:06refactor: remove redundant truncation fallback
- Issue#92akaszubski2026-04-03 01:23feat: model upgrade to Qwen3-Coder (30B + 480B)
- Issue comment#92akaszubski2026-04-03 01:23feat: model upgrade to Qwen3-Coder (30B + 480B)
- Issue#82akaszubski2026-04-03 01:23feat: Claude Code env var configuration for local models
- Issue#93akaszubski2026-04-03 00:45feat: evaluate vllm-mlx as inference server replacement
- Issue#89akaszubski2026-04-03 00:45feat: CLAUDE.md content trimming for local models
- Issue#83akaszubski2026-04-03 00:44feat: tool allowlist for local models
- Issue#81akaszubski2026-02-16 13:43Improve error message quality in backend-client.ts
- Issue#80akaszubski2026-02-16 13:43Improve error message quality in anthropic-proxy.ts
- Issue#79akaszubski2026-02-15 01:58Sync Python/MLX dependencies between development machines
- Issue#78akaszubski2026-02-14 21:36README quickstart: 5-line clone-to-working instructions
- Issue#75akaszubski2026-02-14 21:35First-run wizard: detect missing config and guide setup
- Issue#74akaszubski2026-02-14 18:53security: add request body size limit to prevent OOM
- Issue#71akaszubski2026-02-14 18:53fix: stream error handling hardening (backpressure, keepalive, writableEnded)
- Issue#69akaszubski2026-02-14 04:17Distributed inference across M4 Max + M3 Ultra (pipeline parallel)
- Issue#69akaszubski2026-02-14 00:28Distributed inference across M4 Max + M3 Ultra (pipeline parallel)
- Issue#66akaszubski2026-02-14 00:27Implement proper KV cache warming with stream_generate
- Issue#63akaszubski2026-01-18 13:56test: Expand coverage for critical error paths and untested modules
- Issue#60akaszubski2026-01-18 13:56feat(observability): Add structured JSON logging with request correlation
- Issue#58akaszubski2026-01-17 11:27feat(mlx): add mmap loading for zero-copy KV cache
- Issue#59akaszubski2026-01-17 11:27feat(mlx): add LRU eviction policy for KV cache management
- Issue comment#59akaszubski2026-01-17 11:27feat(mlx): add LRU eviction policy for KV cache management
- Issue#57akaszubski2026-01-17 11:27feat(mlx): add FP8 quantization for KV cache (3x size reduction)
- Issue comment#58akaszubski2026-01-17 11:27feat(mlx): add mmap loading for zero-copy KV cache
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 3 stars here means stars gained during the window, not the repo's star count.