Mistral Small 3.2 24B vs Magistral Small 24B

The two models at a glance

About Mistral Small 3.2 24B

Mistral AI's June 2025 refresh of Small 3.1: a 24B Apache 2.0 dense model with vision input, sharper function calling, and roughly half the rate of runaway generations seen in 3.1. Strengths: Roughly 50% fewer infinite-generation loops than 3.1, Notably improved function calling and structured output reliability, Vision encoder included for multimodal tasks, Apache 2.0 — unrestricted commercial use.

About Magistral Small 24B

Mistral AI's first open reasoning model, built on Small 3.1 with RL-trained chain-of-thought. Hits 70.7% on AIME24 under Apache 2.0. Strengths: First open Mistral reasoner with a real RL training pipeline, AIME24 70.7% — competitive with much larger reasoners, Apache 2.0 license, Runs on a single 24GB GPU at Q4.

How they compare

Mistral Small 3.2 24B comes from Mistral AI and Magistral Small 24B from Mistral AI. This comparison is built entirely from structured specs — parameter count, VRAM by quantization, context window, license, and published benchmark scores — so the verdict below reflects measurable differences rather than marketing claims.

Mistral Small 3.2 24B and Magistral Small 24B share the same 24B parameter class. Both need about 14 GB of VRAM at a Q4 quantization, so they fit the same GPU tier.

The two models target different sweet spots: Mistral Small 3.2 24B is tuned for chat, general, vision, multilingual, fr, while Magistral Small 24B leans toward reasoning, fr. Match the model to your dominant workload rather than to raw size.

Memory, quantization & throughput

Across quantization levels, Mistral Small 3.2 24B requires Q4 ≈ 14 GB, Q5 ≈ 17 GB, Q8 ≈ 26 GB, FP16 ≈ 48 GB, while Magistral Small 24B requires Q4 ≈ 14 GB, Q5 ≈ 17 GB, Q8 ≈ 26 GB, FP16 ≈ 48 GB. In practice Mistral Small 3.2 24B needs a 16 GB card at Q4, so plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity of Q8 or FP16.

Without a GPU, Mistral Small 3.2 24B needs roughly 24 GB of system RAM to run on CPU and Magistral Small 24B about 24 GB — workable for offline use but far slower than GPU inference. On a mid-range GPU you can expect on the order of 15 tokens/sec from Mistral Small 3.2 24B and 15 from Magistral Small 24B, scaling up to 40 and 40 tokens/sec on high-end hardware.

Which fits your GPU

Here is the highest-quality quantization of each model that fits common GPU memory budgets, so you can match Mistral Small 3.2 24B or Magistral Small 24B to the card you actually own:

• On a 16 GB GPU: Mistral Small 3.2 24B runs at Q4 (14 GB); Magistral Small 24B runs at Q4 (14 GB).
• On a 24 GB GPU: Mistral Small 3.2 24B runs at Q5 (17 GB); Magistral Small 24B runs at Q5 (17 GB).

Benchmark scores

Reported benchmarks for Magistral Small 24B: AIME 2024 70.7, MATH-500 90.

Bottom line: which should you pick?

• Pick Mistral Small 3.2 24B if your workload is chat, general, multilingual, vision.
• Pick Magistral Small 24B if your workload is reasoning.

Which GPU should you buy to run Mistral Small 3.2 24B?

To run Mistral Small 3.2 24B locally at Q4, you need ~14 GB of VRAM. The best value for this is a RTX 5070 Ti (16 GB VRAM).

Frequently asked questions

What is the difference between Mistral Small 3.2 24B and Magistral Small 24B?

The headline differences: both are 24B models. Below we break down VRAM by quantization, benchmark scores, and a use-case verdict so you can pick the right one.

Can Mistral Small 3.2 24B and Magistral Small 24B run on a 24 GB GPU?

At a Q4 quantization, Mistral Small 3.2 24B needs about 14 GB of VRAM and fits comfortably on a 24 GB GPU; Magistral Small 24B needs about 14 GB and fits comfortably on a 24 GB GPU. Both have the same Q4 footprint.

What licenses do Mistral Small 3.2 24B and Magistral Small 24B use?

Mistral Small 3.2 24B is licensed under Apache 2.0 and Magistral Small 24B under Apache 2.0.