Mistral Small 3 (24B) — VRAM, Speed & Local Setup

Autor: Jakub Rusinowski · Ostatnia aktualizacja: 28 stycznia 2025

Model library → Mistral Family → Mistral Small 3 (24B)

The new standard for 24GB GPUs. Highly efficient and dense.

Mistral Small 3 (24B) needs about 15 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.

Specifications

Parameters24 Billion
Context window32,000
ArchitectureDense
ProviderMistral AI
LicenceMistral Research
Specified atQ4_K_M
System RAM32 GB
Record updated2025-01-28

Licence

Apache-2.0 — commercial use permitted. Commercial use permitted. No usage restrictions beyond attribution.

VRAM and Speed by Quantization

Modelled on a reference NVIDIA RTX 4090 (24 GB). Assumes an 8K-token context with an f16 KV cache. A longer window needs more; a quantized KV cache needs less. Speed figures are ESTIMATES from the memory-bandwidth roofline described on the methodology page, not benchmarks we ran — rows marked measured come from published or reader-submitted runs. VRAM here includes the KV cache, so it reads higher than the headline figure above, which does not.

QuantBits/weightWeightsVRAM neededEst. speedFit on 24 GB
Q2_K2.637.9 GB10.6 GB~72 tok/s (est.)Fits comfortably
Q3_K_M3.4110.2 GB12.9 GB~59 tok/s (est.)Fits comfortably
Q4_K_M4.8314.5 GB17.2 GB~44 tok/s (est.)Fits comfortably
Q5_K_M5.6717 GB19.7 GB~38 tok/s (est.)Fits comfortably
Q6_K6.5619.7 GB22.4 GB~34 tok/s (est.)Tight fit
Q8_08.5025.5 GB28.2 GB~4 tok/s (est.)Offloads to system RAM (slow)
F1616.0048 GB50.7 GB~2 tok/s (est.)Offloads to system RAM (slow)

Want to set your own context length and KV-cache quantization? Use the interactive VRAM calculator.

Buy This HardwareAMD Radeon RX 9060 XT 16GB — 16 GB VRAM · 160 W board powerDeploy in the Cloud NowRTX 4090 on RunPod — from $0.34/hr · rate checked 2026-07

or compare on Vast.ai from $0.35/hr (typical low · varies)

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

Recommended GPU

The cheapest catalogued GPU that runs Mistral Small 3 (24B) is the AMD Radeon RX 9060 XT 16GB (16 GB).

Ujawnienie afiliacyjne: Niektóre odnośniki na tej stronie to linki afiliacyjne — jeśli dokonasz zakupu za ich pośrednictwem, LLM Configurator może otrzymać prowizję bez dodatkowych kosztów dla Ciebie. Jako uczestnik programu Amazon Associates, LLM Configurator zarabia na kwalifikujących się zakupach.
AMD Radeon RX 9060 XT 16GB
16 GB VRAM · 160 W board power
Ceny w 2026 są niestabilne — sprawdź aktualną ofertę.
Sprawdź cenę na Amazon

How to Run Mistral Small 3 (24B)

Install Ollama, then run:

ollama run mistral-small

Weights on Hugging Face: mistralai/Mistral-Small-24B-Instruct-2501.

Best for: chat, rag.

Can I Run Mistral Small 3 (24B) on My GPU?

Other Mistral Family Sizes

Mistral Small 3 (24B) — Frequently Asked Questions

How much VRAM does Mistral Small 3 (24B) need?
About 15 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does Mistral Small 3 (24B) run on an RTX 4090 (24 GB)?
Yes. Mistral Small 3 (24B) needs about 15 GB at Q4_K_M, inside a 24 GB card, at an estimated 44 tokens/sec.
How do I run Mistral Small 3 (24B) locally?
Install Ollama and run `ollama run mistral-small`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.
What other sizes does Mistral Family come in?
Mistral Small 3 (24B) (15 GB), Mistral NeMo 12B (8 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.

← All Mistral Family models | VRAM calculator | Build a PC for this model | Check your own hardware