Mistral Small 3.2 24B Instruct 2506 — Hardware Requirements & GPU Compatibility
Chat
Mistral Small 3.2 24B Instruct 2506 is a 24.0B-parameter open language model from Mistral AI in the Mistral family. It supports a context window of up to 131,072 tokens. At BF16 it needs about 48.74 GB of VRAM — see which GPUs and Macs can run it below.
592.6K downloads
593 likesSpecifications
- Publisher
- Mistral AI
- Family
- Mistral
- Parameters
- 24.0B
- Architecture
- Mistral3ForConditionalGeneration
- Context Length
- 131,072 tokens
- Vocabulary Size
- 131,072
- License
- Apache 2.0
Get Started
How Much VRAM Does Mistral Small 3.2 24B Instruct 2506 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 48.7 GB | 75.2 GB | 48.02 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run Mistral Small 3.2 24B Instruct 2506?
BF16 · 48.7 GBMistral Small 3.2 24B Instruct 2506 (BF16) requires 48.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 64+ GB is recommended. Using the full 131K context window can add up to 26.4 GB, bringing total usage to 75.2 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Mistral Small 3.2 24B Instruct 2506?
BF16 · 48.7 GB8 devices with unified memory can run Mistral Small 3.2 24B Instruct 2506, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio M4 Max (64 GB).
Runs great
— Plenty of headroom
Decent
— Enough memory, may be tightRelated Models
Frequently Asked Questions
- How much VRAM does Mistral Small 3.2 24B Instruct 2506 need?
Mistral Small 3.2 24B Instruct 2506 requires 48.7 GB of VRAM at BF16. Full 131K context adds up to 26.4 GB (75.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 24.0B × 16 bits ÷ 8 = 48 GB
KV Cache + Overhead ≈ 0.7 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 27.2 GB (at full 131K context)
VRAM usage by quantization
BF1648.7 GBBF16 + full context75.2 GB- Can NVIDIA GeForce RTX 5090 run Mistral Small 3.2 24B Instruct 2506?
No — Mistral Small 3.2 24B Instruct 2506 requires at least 48.7 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Mistral Small 3.2 24B Instruct 2506 on a Mac?
Mistral Small 3.2 24B Instruct 2506 requires at least 48.7 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Mistral Small 3.2 24B Instruct 2506 locally?
Yes — Mistral Small 3.2 24B Instruct 2506 can run locally on consumer hardware. At BF16 quantization it needs 48.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Mistral Small 3.2 24B Instruct 2506?
At BF16, Mistral Small 3.2 24B Instruct 2506 can reach ~60 tok/s on AMD Instinct MI300X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 48.7 × 0.55 = ~60 tok/s
Estimated speed at BF16 (48.7 GB)
~60 tok/s~45 tok/s~37 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Mistral Small 3.2 24B Instruct 2506?
At BF16, the download is about 48.02 GB.
- Which GPUs can run Mistral Small 3.2 24B Instruct 2506?
No single consumer GPU has enough VRAM to run Mistral Small 3.2 24B Instruct 2506 at BF16 (48.7 GB). Multi-GPU or professional hardware is required.
- Which devices can run Mistral Small 3.2 24B Instruct 2506?
8 devices with unified memory can run Mistral Small 3.2 24B Instruct 2506 at BF16 (48.7 GB), including Mac Pro M2 Ultra (192 GB), Mac Studio M2 Ultra (192 GB), Mac Studio M4 Max (128 GB), Mac Studio M4 Max (64 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.