Llama 3.2 1B Instruct › Apple M3 Ultra (80-core GPU)
Llama 3.2 1B Instruct on the Apple M3 Ultra (80-core GPU)
Best measured: 178 tokens/s generation and 5,300 tokens/s prompt processing (Q4_K_M, llamafile).
For comparison, our size-scaling estimate for this combination is 447 tok/s.
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| Q4_K_M | llamafile | 179 | 5,601 | LocalScore avg | LocalScore rank 75/878 | LocalScore (localscore.ai) |
| Q4_K_M | llamafile | 178 | 5,000 | LocalScore avg | LocalScore rank 83/878 | LocalScore (localscore.ai) |
tg128 = generating 128 tokens; pp512 = reading a 512-token prompt (llama-bench). LocalScore averages several prompt and output lengths, so its numbers run lower than tg128.
No card yet?
Try it on a rented GPU (RunPod)Affiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.
All results for the Apple M3 Ultra (80-core GPU) · Llama 3.2 1B Instruct on other hardware