llamaperf
Sep 7, 2026
Throughput
114.2 t/s gen
Quant
Q6_K (GGUF)
VRAM reported
96 GB

Use cases

coding

Summary

User compares Qwen 3.8 27B community quants on an RTX Pro 6000 96GB, with atomic ad-q6_k averaging 114.17 t/s as the best. The quants tested are atomic ad-q6_k, unsloth ud-q6_k_l and bartowski q6_k. The user also compares the results against a Claude Opus 4.6 subscription, with prompts generating self-playing 3D HTML games.