llamaperf
Submit

Qwen3.8 27B

on 2× NVIDIA RTX 3060 12GB · llama.cpp · 131,072 ctx

NVIDIA hardware
Sep 7, 2026
Throughput
50.6 t/s gen · 573.0 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.
Quant
Q4_K_XL (GGUF)
KV cache
q8_0
MTP (Multi-Token Prediction)
on
VRAM reported
24 GB

Summary

User reports a dual RTX 3060 12GB setup with tensor split. MTP draft accept is 4/4 (100%) on probe. VRAM usage is 23.2/24.0 GiB.
View source ↗