- Hero verified-stats one-liner at top (256K, 31/31, 200K needle 3/3, 9/9 e2e, $4.12)
- Reframe Status: HF Spaces don't ship MI300X by default → CPU mock is by design,
not because the project doesn't work. Verified numbers come from real MI300X
stress test on AMD Developer Cloud (124 min, $4.12)
- Add demo video link (youtu.be/BvSBR1QazLU)
- Add Lablab project page + AMD Developer Forum thread #505 + GitHub links
- Memory-architecture comparison as table (MI300X vs H100 80GB)
- Soften like-CTA: factual rather than pleading
- README.md: parallel updates for Space card display
fix: links iframe-safe — switch markdown links to HTML <a target=_blank>; footer uses gr.HTML for raw anchor support
Open-source coding agent that ingests an entire git repository and reasons
across the whole codebase with multi-step tool use. Built for the AMD
Developer Hackathon 2026.
This Space ships:
- Ingestion pipeline (clone + tree-sitter + smart chunker + priority budget)
- 5 tools: read_file, grep_codebase, execute_code, run_tests, git_log
- SC-TIR-style agent loop adapted from AIMO3 math reasoning pipeline
- Mock LLM backend for offline demo (default)
- vLLM client ready to wire to MI300X (Qwen3-Coder-Next-FP8, 256K context)
Full source: https://github.com/SRKRZ23/repomind