Mistral on NVIDIA Ada
Can I run Mistral Small 24B on an RTX 4080 Super?
Not at Q4_K_M — it needs 15.5 GB against 14.2 GB available. Drop to IQ4_XS and it fits, at about 35.1 tokens per second.
Won't fit
15.5 GB of 14.2 GB · 109%016 GB
Weights
13.4 GB
KV cache
1.3 GB
Runtime overhead
0.8 GB
Over the limit
1.3 GB
Short by 1.3 GB. You can run it with 36 of 40 layers on the RTX 4080 Super and the rest in system RAM, at roughly 15.3 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.
Generation15.3tok/s
Prompt processing927tok/s
Max context0tokens
KV per 1K tokens0GB
Every quantisation of Mistral Small 24B on a RTX 4080 Super
Highlighted row is the highest quality that still fits at 8K context.
| Quantisation | bpw | Weights | Total | Verdict | Max ctx | tok/s |
|---|---|---|---|---|---|---|
| FP16 / BF16 | 16.00 | 43.9 GB | 46.0 GB | Won't fit | — | 1.15 |
| INT8 / W8A8 | 8.50 | 23.0 GB | 25.1 GB | Won't fit | — | 3.11 |
| Q8_0 (GGUF) | 8.50 | 23.0 GB | 25.1 GB | Won't fit | — | 3.11 |
| FP8 (E4M3) | 8.00 | 22.0 GB | 24.0 GB | Won't fit | — | 3.40 |
| Q6_K | 6.56 | 18.0 GB | 20.1 GB | Won't fit | — | 5.35 |
| Q5_K_M | 5.67 | 15.6 GB | 17.6 GB | Won't fit | — | 8.08 |
| Q5_K_S | 5.52 | 15.1 GB | 17.2 GB | Won't fit | — | 8.99 |
| AWQ 4-bit | 4.25 | 13.5 GB | 15.6 GB | Won't fit | — | 15.2 |
| GPTQ 4-bit | 4.25 | 13.5 GB | 15.6 GB | Won't fit | — | 15.2 |
| MXFP4 | 4.25 | 13.5 GB | 15.6 GB | Won't fit | — | 15.2 |
| Q4_K_M | 4.85 | 13.4 GB | 15.5 GB | Won't fit | — | 15.3 |
| Q4_K_S | 4.58 | 12.7 GB | 14.8 GB | Won't fit | 4K | 21.6 |
| Q4_0 | 4.55 | 12.6 GB | 14.7 GB | Won't fit | 5K | 21.7 |
| IQ4_XS | 4.25 | 11.9 GB | 13.9 GB | Fits, but tight | 10K | 35.1 |
| Q3_K_M | 3.91 | 11.0 GB | 13.1 GB | Fits, but tight | 16K | 37.7 |
| IQ3_M | 3.70 | 10.4 GB | 12.5 GB | Runs comfortably | 19K | 39.5 |
| IQ3_XXS | 3.06 | 8.8 GB | 10.9 GB | Runs comfortably | 30K | 46.4 |
| Q2_K | 2.63 | 7.7 GB | 9.8 GB | Runs comfortably | 32K | 52.6 |
| IQ2_XXS | 2.06 | 6.2 GB | 8.3 GB | Runs comfortably | 32K | 63.8 |
| IQ1_M | 1.75 | 5.4 GB | 7.5 GB | Runs comfortably | 32K | 72.1 |