Builds / Apple M5 Max (unified)
Mac Studio M5 Max 36 GB: Qwen3.6 35B-A3B at 173 tok/s (Rapid-MLX), 146 on oMLX
Source: rapidmlx.com ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
- GPU(s)
- Apple M5 Max (unified)
- VRAM / unified memory (GB)
- 36
- CPU
- Apple M5 Max
- Motherboard
- not stated
- RAM
- 36 GB unified
- PSU
- not stated
- Other
- Mac Studio; Rapid-MLX 0.16.1, oMLX 0.7.0, LM Studio llmster 0.0.25 (MLX runtime 1.11.0); 4-bit MLX weights
- Reported total cost
- not stated
- Power
- not stated
Measured speed (generation tokens/sec)
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Qwen3.6 35B-A3B | 4-bit MLX | Rapid-MLX 0.16.1 (MTP) | 173.4 | ⚡ spec-decode (MTP) |
| Qwen3.6 35B-A3B | 4-bit MLX | oMLX 0.7.0 | 145.8 | |
| Qwen3.6 35B-A3B | 4-bit MLX | LM Studio (MLX) | 124.5 | |
| Qwen3.8 27B | 4-bit MLX | Rapid-MLX 0.16.1 (MTP) | 37.9 | ⚡ spec-decode (MTP) |
| Qwen3.8 27B | 4-bit MLX | oMLX 0.7.0 | 25.3 | |
| Qwen3.8 27B | 4-bit MLX | LM Studio (MLX) | 26.0 | |
| Qwen3.5 9B | 4-bit MLX | Rapid-MLX 0.16.1 (MTP) | 115.5 | ⚡ spec-decode (MTP) |
| Qwen3.5 9B | 4-bit MLX | oMLX 0.7.0 | 78.0 |
Measured 12 Oct 2026, single-stream decode, single run per engine. Published by the Rapid-MLX developers (vendor benchmark). 4-stream aggregate: Qwen3.6 35B-A3B 293.9 tok/s (Rapid-MLX), Qwen3.8 27B 82.8 tok/s.
Source: https://rapidmlx.com/compare
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)