MacBook Pro M5 Max 128 GB: Qwen3.6 35B-A3B at 271 tok/s (mlx-serve + MTP), Gemma 4 26B-A4B at 166
| GPU(s) | Apple M5 Max (unified) |
|---|---|
| VRAM / unified memory (GB) | 128 |
| CPU | Apple M5 Max |
| Motherboard | |
| RAM | 128 GB unified |
| PSU | |
| Other | MacBook Pro on AC power, macOS 26.6; mlx_lm.server 0.32.2, Rapid-MLX 0.16.1, mlx-serve 26.10.2; 4-bit MLX weights |
| Reported total cost | |
| Power |
Measured speed (generation tokens/sec)
📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Qwen3.6 35B-A3B | 4-bit MLX (mlx-serve build) | mlx-serve 26.10.2 + MTP | 271 | ⚡ spec-decode (MTP) |
| Qwen3.6 35B-A3B | 4-bit MLX | Rapid-MLX 0.16.1 + MTP | 240 | ⚡ spec-decode (MTP) |
| Qwen3.6 35B-A3B | 4-bit MLX | mlx_lm.server 0.32.2 | 107 | |
| Gemma 4 26B-A4B | 4-bit MLX (QAT) | Rapid-MLX 0.16.1 + DFlash | 166 | ⚡ spec-decode (DFlash) |
| Qwen3.8 27B | 4-bit MLX | same server, no speculation | 31 |
Measured 19 Oct 2026, temperature 0, one run per setup (repeats varied up to 15%). Qwen3.6 figures are on the coding prompt. Qwen3.8 27B ran about 12 tok/s on battery vs about 31 on AC. 4 concurrent requests: 412 tok/s total on mlx-serve + MTP.
Source: https://baobabtech.ai/posts/local-moe-m5-max-qwen-gemma-speed
Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.
Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)