nvfp4-vs-fp8-kv-cache-terminal-bench
Python
★ 2
updated 11d ago
FP8 vs NVFP4 (4-bit) KV cache on Terminal-Bench 2.0 — no measurable accuracy loss, 1.78x more KV capacity. Full results + verify.py.
No plain-English explanation yet — one is being written right now. Check back in a minute.