Benchmark result
ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:Q2_0 on Apple M4 Max — 90.7 tok/s
Measured with vLLM on November 11, 2026.
Support llm-bench.io
llm-bench.io is free. If it helped you, I'd appreciate your support to cover the infrastructure costs.
What the model built
Open full screen →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
1st fastest of 2 runs of this model on Apple M4 Max · median 74.8 tok/s.
sort
- 190.7tok/s55.3vLLM · Q2_0this run
- 258.8tok/s86.1vLLM · Q2_0
ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:Q2_0 on other hardware
Reproduce this run
toolvLLM
modelISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:Q2_0 (Q2_0)
context8,192
thinkingoff
clientv0.4.99+102
Set those in vLLM, then:
llm-benchmark benchmark --model "ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:Q2_0" --tool "vLLM"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.