Magistral Small 1.2 (2509)
Magistral Small 1.2 (2509) by Mistral AI (24B parameters). At Q4_K_M with an 8K context it needs about 16.0 GiB of memory. Mistral's 24B reasoning model with image input; it fits a 16 GB card at 4-bit. Ollama's magistral tag is the older 2506 build, not this one.
Last updated: 2026-10-13
Model facts
Memory needed
Weights + KV cache + runtime overhead, in GiB. Lower quantization = less memory, slightly lower quality.
| Quantization | 4K context | 8K context | 32K context |
|---|---|---|---|
| Q4_K_M | 15.4 GiB | 16.0 GiB | 19.7 GiB |
| Q5_K_M | 17.8 GiB | 18.4 GiB | 22.2 GiB |
| Q6_K | 20.5 GiB | 21.1 GiB | 24.8 GiB |
| Q8_0 | 26.1 GiB | 26.7 GiB | 30.4 GiB |
| FP16 | 48.1 GiB | 48.7 GiB | 52.4 GiB |
Which hardware runs it
Best quantization that fits each device with a 8K context. Estimates, not benchmarks. See the methodology.
Frequently asked questions
How much VRAM does Magistral Small 1.2 (2509) need?
With an 8K context, about 16.0 GiB at Q4_K_M and 26.7 GiB at Q8_0 (weights + KV cache + runtime overhead). Longer contexts need more memory for the KV cache.
Can Magistral Small 1.2 (2509) run on an GeForce RTX 4090?
Yes. Magistral Small 1.2 (2509) fits on the GeForce RTX 4090 (24 GB) at Q6_K with an 8K context, using about 21.1 GiB, at an estimated 36 tokens/s.
Can I run Magistral Small 1.2 (2509) on a Mac?
Yes. Magistral Small 1.2 (2509) fits on the M3 Ultra (96 GB) at FP16 with an 8K context, using about 48.7 GiB, at an estimated 11 tokens/s.
Can Magistral Small 1.2 (2509) be used commercially?
Magistral Small 1.2 (2509) is released under Apache-2.0. The licence allows commercial use (read its conditions).