Benchmark result
Muse-Glimmer-30B-Heretic_t1p95k64m05 on 2× Tesla V100-PCIE-32GB — 58.4 tok/s
Measured with llama.cpp manual on November 10, 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 2× Tesla V100-PCIE-32GB · Multi-GPU · median 52.3 tok/s.
sort
- 158.4tok/s76.6llama.cppthis run
- 246.1tok/s80.3llama.cpp
Reproduce this run
toolllama.cpp manual
modelMuse-Glimmer-30B-Heretic_t1p95k64m05
context32,768
think limit8,096 tokens
samplingtemperature 1 · top_p 0.949999988079071 · top_k 64 · min_p 0.05000000074505806 · presence_penalty 0 · repeat_penalty 1
clientv0.4.99+102
Set those in llama.cpp, then:
llm-benchmark benchmark --model "Muse-Glimmer-30B-Heretic_t1p95k64m05" --tool "llama.cpp"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.