Home › Models › Kimi K2.7 Code › Apple M3 Ultra (80-core GPU)
Kimi K2.7 Code on the Apple M3 Ultra (80-core GPU)
Kimi K2.7 Code runs at about 24.6 tokens/s on the Apple M3 Ultra (80-core GPU) (measured, UD-Q2_K_XL, Metal, llama-bench).
UD-Q2_K_XL · Metal
same run
For comparison, our size-scaling estimate for this combination is 24.7 tok/s.
Every run
| Quant | Backend | Generate tok/s | Prompt tok/s | Test | Build / by / notes | Source |
|---|---|---|---|---|---|---|
| UD-Q2_K_XL | Metal | 24.6 | 145 | tg300 (llama-bench -n 300) | 571d0d5 Mac Studio M3 Ultra 512 GB, weights fully in memory, single run. | AI KIZAI: Mac Studio M3 Ultra 512 GB measured benchmarks |
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. "decode, llama-server" = single-stream generation speed measured against a running llama-server (no speculative decoding).
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) · Kimi K2.7 Code on other hardware
Embed this result
HTML
<a href="https://rigbench.pages.dev/model/kimi-k2-7-code/on/apple-m3-ultra-80c/"><img src="https://rigbench.pages.dev/badge/kimi-k2-7-code--apple-m3-ultra-80c.svg" alt="Kimi K2.7 Code · Apple M3 Ultra (80-core GPU): 24.6 tok/s (UD-Q2_K_XL, measured)"></a>
Markdown (GitHub README, forums)
[](https://rigbench.pages.dev/model/kimi-k2-7-code/on/apple-m3-ultra-80c/)