Can Mac M5 Ultra 512G run Mistral Large 4?
Comfortable: GGUF IQ2_XXS needs 311.5 GB at 4K, batch 1; headroom 72.5 GB.
Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.
Measured runs
No measured run for this pair. Sizing is an estimate, not a runtime test.
Quant memory at 4K
| Quant | Total GB | Verdict | Weight sizing |
|---|---|---|---|
| GGUF IQ2_XXS | 311.5 | Comfortable | Verified file |
| GGUF Q2_K | 397.8 | Tight | Verified file |
| GGUF Q4_K_M | 663.8 | Does not fit | Verified file |
| GGUF Q8_0 | 1165.4 | Does not fit | Verified file |
Context capacity · GGUF IQ2_XXS
| Context | Total GB | Verdict |
|---|---|---|
| 4K | 311.5 | Comfortable |
| 16K | 312.6 | Comfortable |
| 32K | 314.0 | Comfortable |
| 128K | 322.4 | Comfortable |
Generation speed
GGUF IQ2_XXS: 86 tok/s · Bandwidth ceiling (not measured)