Long-context faithfulness benchmark + audit-friendly Qwen3 quantized releases. Outputs ship; inputs are auditable.
🏗️ Building on HF
Muhammad Awais
drawais
AI & ML interests
Needle-1M-bench: a long-context faithfulness benchmark on real arxiv papers. Centralized scoring.
Organizations
None yet