| Age | Commit message (Expand) | Author |
|---|
| 6 days | fix: warn before using a likely too-big chunk size for sglang-omni models | historia |
| 6 days | feat: reserve 20% vram when running higgs with sglang-omni | historia |
| 6 days | feat: sglang-omni model picker on manual server launch | historia |
| 6 days | fix: ZONOS2 only generating 1024 tokens max | historia |
| 6 days | feat: sglang backend support | historia |
| 8 days | feat: model capabilities lined up in tui menu | historia |
| 9 days | fix: non-clone models correctly supported in tui, restart server when needed | historia |
| 10 days | fix: crash on bad model_specs from audio.cpp, sanitized | historia |
| 11 days | feat(audio.cpp): if github binary download fails, automatically start buildin... | historia |
| 11 days | feat: offer to download prebuilt mac/win binaries, new macos build fallback i... | historia |
| 11 days | feat: run auto-build audio.cpp scripts for macos and windows | historia |
| 11 days | refactor: overhaul config.py, remove cli default options | historia |
| 11 days | fix: silently failing whisper transcription | historia |
| 12 days | feat: help menu in tui | historia |
| 12 days | feat: menu for language options | historia |
| 12 days | feat: separate venvs for qwen-tts and faster, manage (un)installs | historia |
| 13 days | feat: manage qwen-tts model installs manually, delete model(s) when uninstalled | historia |
| 13 days | feat: design model support for qwen-tts backend. remove unnecessary port spli... | historia |
| 13 days | fix: settings menu only prompts to save after change | historia |
| 13 days | feat: language and option fields in tui, context-sensitive voice label | historia |
| 13 days | feat: combined install/configure tui screens into one menu, removed extraneou... | historia |
| 13 days | refactor: audiocpp.py setup flow | historia |
| 14 days | feat: rename convert to generate, line up model menu options | historia |
| 14 days | fix: build audio.cpp with --deployment-build | historia |
| 14 days | feat: combine --speaker and --voice for clarity | historia |
| 14 days | feat: automatically name audio.cpp model ids with long name | historia |
| 2026-08-25 | feat(tui): simultaneous build and model download | historia |
| 2026-08-24 | feat: user-friendly menu gating, clearer install/configure path for backends | historia |
| 2026-08-24 | feat: configure [backend] loads existing backend config rather than just over... | historia |
| 2026-08-24 | feat: option to unload models from audio.cpp per-run | historia |
| 2026-08-24 | feat: clearer split between local (managed) and remote URLs and server status | historia |
| 2026-08-24 | feat: audio.cpp unloads model before converting | historia |
| 2026-08-24 | fix: don't expect/use local config for remote servers | historia |
| 2026-08-24 | feat: settings for ports in TUI | historia |
| 2026-08-24 | chore: update backend docs with custom venv | historia |
| 2026-08-24 | refactor: add app directory, dir structure change | historia |
