Configs › Qwen3-14B tok/s: Mac Max 62.5, Mac Ultra 55.4
Qwen3-14B tok/s: Mac Max 62.5, Mac Ultra 55.4
Measured Qwen3-14B speeds: 62.5 tok/s on Mac Max (MLX-4BIT, mlx); 55.4 tok/s on Mac Ultra. 3 configs, each linked to its source, snapshot 2026-10-15.
Snapshot 2026-10-15 · 394 configs on AI boxes and 423 on GPU cards · 123 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
Quick answers
- How fast is Qwen3-14B on Mac Max?
- 62.5 tok/s single-stream decode (MLX-4BIT, mlx), the fastest of 2 measured configs on Apple M-series Max (MacBook Pro / Mac Studio Max), measured by emihiggins, source last checked 2026-09-29. source
- How fast is Qwen3-14B on Mac Ultra?
- 55.4 tok/s single-stream decode (MLX-4BIT, mlx), the only measured config on Apple Mac Studio (M-series Ultra), measured by theNetworkChuck, source last checked 2026-09-29. source
Measured configs for Qwen3-14B
| Model | Hardware | Quant | Backend | Decode | Context | Trust | Source |
|---|---|---|---|---|---|---|---|
| Qwen3-14B | Mac Max | MLX-4BIT | mlx | 62.5 tok/s (prefill 1739) | 512 | Extracted, source-linked | source |
| Qwen3-14B | Mac Max | Q4_K_M | llama.cpp | 57.4 tok/s (prefill 1584) | 512 | Extracted, source-linked | source |
| Qwen3-14B | Mac Ultra | MLX-4BIT | mlx | 55.4 tok/s (prefill 2278) | 32000 | Extracted, source-linked | source |
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.