Best GPU for Mistral-Small-24B-Instruct-2501
The smallest memory budget in our catalog that runs Mistral-Small-24B-Instruct-2501 cleanly at 8,192 tokens is 18 GiB net, with Q4_K_M (14.7 GiB–15.5 GiB total). Cards at that budget: Radeon RX 7900 XT 20GB. The largest file, BF16, is comfortable from 78 GiB net: NVIDIA A100 80GB SXM, NVIDIA H100 80GB SXM.
Which card, by net memory
| Net budget | Devices | Best file that runs | Total at 8,192 | Result | Same file at 32,768 |
|---|---|---|---|---|---|
| 15 GiB | GeForce RTX 4060 Ti 16GB, GeForce RTX 4070 Ti SUPER 16GB | Q4_K_M | 14.7 GiB–15.5 GiB | Tight | No |
| 18 GiB | Radeon RX 7900 XT 20GB | Q4_K_M | 14.7 GiB–15.5 GiB | Fits | No |
| 22 GiB | GeForce RTX 3090 24GB, GeForce RTX 4090 24GB, Radeon RX 7900 XTX 24GB | Q5_K_M | 17.0 GiB–17.8 GiB | Fits | Tight |
| 30 GiB | GeForce RTX 5090 32GB | Q8_0 | 24.7 GiB–25.6 GiB | Fits | Tight |
| 46 GiB | NVIDIA L40S 48GB | Q8_0 | 24.7 GiB–25.6 GiB | Comfortable | Comfortable |
| 54 GiB | Apple M2 Max 64GB unified (MacBook Pro / Mac Studio) | BF16 | 45.8 GiB–47.7 GiB | Fits | Tight |
| 78 GiB | NVIDIA A100 80GB SXM, NVIDIA H100 80GB SXM | BF16 | 45.8 GiB–47.7 GiB | Comfortable | Comfortable |
Rows start at the first budget where the smallest official file is at least tight and stop at the first where the largest file is comfortable. Devices in one row have the same net budget (capacity minus the catalog's default reserve), so they get the same verdict; choose between them on bandwidth, software support and price, which this page does not rank. llama.cpp, one sequence, fp16 KV cache, typical extras. Unified-memory devices use the reserve in their catalog record. "n/a" in the last column means the model's native window is shorter than 32,768 tokens.
Files this ranking uses
| File | Weights | Source repository |
|---|---|---|
| BF16 | 43.91 GiB (47,144,806,400 bytes) | mistralai/Mistral-Small-24B-Instruct-2501 |
| Q8_0 | 23.33 GiB (25,054,779,648 bytes) | unsloth/Mistral-Small-24B-Instruct-2501-GGUF |
| Q6_K | 18.02 GiB (19,345,938,688 bytes) | unsloth/Mistral-Small-24B-Instruct-2501-GGUF |
| Q5_K_M | 15.61 GiB (16,763,983,552 bytes) | bartowski/Mistral-Small-24B-Instruct-2501-GGUF |
| Q4_K_M | 13.35 GiB (14,333,909,248 bytes) | unsloth/Mistral-Small-24B-Instruct-2501-GGUF |
Byte counts come from the repository file listings recorded on the model page. Memory estimates are formulas, not measurements on these cards. No speed figure is given: none was measured here.
Data sources
- Mistral-Small-24B-Instruct-2501 model card
- config.json — revision 9527884be6e5616bdd54de542f9ae13384489724
- safetensors index — metadata.total_size=47144806400
- Method and reserves — catalog dataset 2026-10-02.1; hardware capacities from vendor spec pages listed on each device page