Hardware1
Apple M5
4 runsBest decode · TG128
172.7tok/s
Best prefill · PP512
4,997tok/s
Badges3
Record1
Pioneer1
Coverage1
4 decode records
Fastest result in a tested configuration
Record
First run
Submitted a benchmark report
Pioneer
Model explorer
4 models covered
Coverage
Submissions4
Sort by
Order
Prefill size
4 submissions · PP512 selected
| Model / runtime / format | Device | Backend | Report | |||
|---|---|---|---|---|---|---|
ornith-ai/Ornith-1.5-9B BaseRTQ4· default-q4 This report predates artifact identity, so ComputeArena mapped its model name to a model family by hand. The exact model bytes were not verified. | Apple M5 | Metal | 14.9 tok/s TG128 | 281 tok/s PP512 | ||
basecompute/gemma-4-E2B-it BaseRTQ4· default-q4 The report's model SHA-256 matches a file published on Hugging Face, so the exact model bytes are known. This does not attest benchmark execution. | Apple M5 | Metal | 48.7 tok/s TG128 | 3,484 tok/s PP512 | ||
basecompute/Qwen3.5-2B BaseRTQ4· default-q4 The report's model SHA-256 matches a file published on Hugging Face, so the exact model bytes are known. This does not attest benchmark execution. | Apple M5 | Metal | 58.6 tok/s TG128 | 1,203 tok/s PP512 | ||
Qwen/Qwen3-0.6B BaseRTQ4· default-q4 This report predates artifact identity, so ComputeArena mapped its model name to a model family by hand. The exact model bytes were not verified. | Apple M5 | Metal | 172.7 tok/s TG128 | 4,997 tok/s PP512 |