Benchmark result

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_S-mtp on Apple M4 Pro — 10.7 tok/s

Measured with llama.cpp (MTP) manual on October 15, 2026.

What the model built

The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.

How this run compares

ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_S-mtp on other hardware

32.2tok/s79.5Advanced Micro Devices, Inc. [AMD/ATI] Device 7551

Reproduce this run

toolllama.cpp (MTP) manual

modelISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_S-mtp

context262,144

thinkingon

clientv0.4.93+102

Set those in llama.cpp (MTP), then:

llm-benchmark benchmark --model "ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_S-mtp" --tool "llama.cpp (MTP)"

The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.