Till JS c3c02c6a22 feat(manacore/web): add benchmark, compare history, markdown & cache status to LLM test ... - Benchmark tab: run same prompt N times against loaded model, show avg/min/max/median for latency and tok/s with per-run table and bars - Compare history: save results to localStorage (max 20), restore or delete previous comparisons via History panel - Markdown rendering: use marked for assistant responses in chat tab and compare outputs with proper code blocks, lists, headings - Model cache status: check browser Cache API via WebLLM hasModelInCache, show cached/uncached per model with size badges - Add marked dependency Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>		2026-04-02 11:50:23 +02:00
..
landing	chore: remove playground, reader, bauntown, voxelava, and worldream apps	2026-04-02 11:47:54 +02:00
mobile	refactor: rename ManaDeck to Cards across entire monorepo	2026-04-01 11:45:21 +02:00
web	feat(manacore/web): add benchmark, compare history, markdown & cache status to LLM test	2026-04-02 11:50:23 +02:00