No — not on this device
Mistral Small 4 119B-A6B at Q4_K_M needs 67.7 GB against 51.2 GB usable, and the shortfall of 16.5 GB is more than 64 GB of system RAM can cover at a tolerable speed. A smaller sibling or a lower quantisation is the honest answer here.
Instruct, reasoning, vision and code in one 119B MoE. A latent KV cache (320 wide) keeps context cheap; the weights still want 64 GB+.
What hardware do I need for Mistral Small 4 119B-A6B? →
Fits instead: Mistral Small 3.2 24B (15.1 GB)
The VRAM budget
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 117.8 GB | 118.5 GB | — | ~2.6 | −0.1% ppl | 67.3 GB over |
| Q6_K | 90.9 GB | 91.7 GB | — | ~3.4 | −0.4% ppl | 40.5 GB over |
| Q5_K_M | 78.5 GB | 79.3 GB | — | ~3.9 | −0.8% ppl | 28.1 GB over |
| Q4_K_M | 66.9 GB | 67.7 GB | — | ~4.6 | −1.9% ppl | 16.5 GB over |
| Q3_K_M | 54.2 GB | 54.9 GB | — | ~5.7 | −5.4% ppl | 3.7 GB over |
| Q2_K | 46.4 GB | 47.2 GB | 190K | 6.6 | −15% ppl | Long context |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. This model uses multi-head latent attention, so its cache is a compressed latent rather than full K and V.
How to run it
$ llama-server \
-hf mistralai/Mistral-Small-4-119B-2603:Q4_K_M \
-c 8192 -ngl 27
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.