Benchmark result
google/gemma-4-e4b on Apple M1 Max — 45.8 tok/s
Measured with LM Studio on November 4, 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
4th fastest of 6 runs of this model on Apple M1 Max · median 46.4 tok/s.
sort
- 148.3tok/s66.7LM Studio · 4bit
- 248.0tok/s62.9LM Studio · 4bit
- 347.0tok/s60.9LM Studio · 4bit
- 445.8tok/s60.4LM Studio · Q4_K_Mthis run
- 545.5tok/s60.6LM Studio · 4bit
- 645.3tok/s69.0LM Studio · 4bit
Reproduce this run
toolLM Studio
modelgoogle/gemma-4-e4b (Q4_K_M)
context32,768
thinkingon
clientv0.4.98+102
Set those in LM Studio, then:
llm-benchmark benchmark --model "google/gemma-4-e4b" --tool "LM Studio"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.