Apple M1 (7-core GPU)

15measured results

38.9best gen tok/s*

8 GBreported memory

*Headline excludes tiny/test models (under 0.6B params or TinyStories-style). Open in rig checker →

llama.cpp

Model	Quant	Gen tok/s	Prompt tok/s	Source	Context tested	Build
Llama 3.2 1B Instruct	Q4_K - Medium	38.9	537	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	38.6	536	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	38.5	544	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	38.4	540	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	38.4	528	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	38.3	535	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	37.3	509	LocalScore ↗	4,352	llamafile 0.9.2	Details →
Llama 3.2 1B Instruct	Q4_K - Medium	36.0	503	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 2 7B	Q4_0	14.2	108	llama.cpp (Metal) ↗	512+128	llama.cpp Metal	Details →
Meta Llama 3.1 8B Instruct	Q4_K - Medium	13.4	109	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Meta Llama 3.1 8B Instruct	Q4_K - Medium	9.7	88.0	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Meta Llama 3.1 8B Instruct	Q4_K - Medium	9.1	79.9	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Llama 2 7B	Q8_0	7.9	108	llama.cpp (Metal) ↗	512+128	llama.cpp Metal	Details →
Qwen2.5 14B Instruct	Q4_K - Medium	4.8	40.7	LocalScore ↗	512+128	llamafile 0.9.2	Details →
Qwen2.5 14B Instruct	Q4_K - Medium	4.8	40.6	LocalScore ↗	512+128	llamafile 0.9.2	Details →

Ollama

No measured Ollama results yet. Submit yours →

MLX

No measured MLX results yet. Submit yours →

Includes results from LocalScore, a Mozilla Builders project by cjpais, mirrored in the cn0303/local-llm-speed-benchmarks dataset. LocalScore runs use llamafile, which is built on llama.cpp.