Mistral Small 24B Instruct 2501 Quantized.w8a8 — Hardware Requirements & GPU Compatibility
Chat
Mistral Small 24B Instruct 2501 Quantized.w8a8 is a 23.6B-parameter open language model from RedHatAI in the Mistral family. It supports a context window of up to 32,768 tokens. At BF16 it needs about 47.87 GB of VRAM — see which GPUs and Macs can run it below.
15.7K downloads
2 likesSpecifications
- Publisher
- RedHatAI
- Family
- Mistral
- Parameters
- 23.6B
- Architecture
- MistralForCausalLM
- Context Length
- 32,768 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2025-10-29
- License
- Apache 2.0
Get Started
How Much VRAM Does Mistral Small 24B Instruct 2501 Quantized.w8a8 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 47.9 GB | 54.2 GB | 47.15 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run Mistral Small 24B Instruct 2501 Quantized.w8a8?
BF16 · 47.9 GBMistral Small 24B Instruct 2501 Quantized.w8a8 (BF16) requires 47.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 63+ GB is recommended. Using the full 33K context window can add up to 6.3 GB, bringing total usage to 54.2 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run Mistral Small 24B Instruct 2501 Quantized.w8a8?
BF16 · 47.9 GB11 devices with unified memory can run Mistral Small 24B Instruct 2501 Quantized.w8a8, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Studio M4 Max (64 GB).
Runs great
— Plenty of headroom
Related Models
Frequently Asked Questions
- How much VRAM does Mistral Small 24B Instruct 2501 Quantized.w8a8 need?
Mistral Small 24B Instruct 2501 Quantized.w8a8 requires 47.9 GB of VRAM at BF16. Full 33K context adds up to 6.3 GB (54.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 23.6B × 16 bits ÷ 8 = 47.2 GB
KV Cache + Overhead ≈ 0.7 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 7 GB (at full 33K context)
VRAM usage by quantization
BF1647.9 GBBF16 + full context54.2 GB- Can NVIDIA GeForce RTX 5090 run Mistral Small 24B Instruct 2501 Quantized.w8a8?
No — Mistral Small 24B Instruct 2501 Quantized.w8a8 requires at least 47.9 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run Mistral Small 24B Instruct 2501 Quantized.w8a8 on a Mac?
Mistral Small 24B Instruct 2501 Quantized.w8a8 requires at least 47.9 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 24B Instruct 2501 Quantized.w8a8 locally?
Yes — Mistral Small 24B Instruct 2501 Quantized.w8a8 can run locally on consumer hardware. At BF16 quantization it needs 47.9 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Mistral Small 24B Instruct 2501 Quantized.w8a8?
At BF16, Mistral Small 24B Instruct 2501 Quantized.w8a8 can reach ~61 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 ÷ 47.9 × 0.55 = ~61 tok/s
Estimated speed at BF16 (47.9 GB)
~61 tok/s~46 tok/s~38 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 24B Instruct 2501 Quantized.w8a8?
At BF16, the download is about 47.15 GB.
- Which GPUs can run Mistral Small 24B Instruct 2501 Quantized.w8a8?
No single consumer GPU has enough VRAM to run Mistral Small 24B Instruct 2501 Quantized.w8a8 at BF16 (47.9 GB). Multi-GPU or professional hardware is required.
- Which devices can run Mistral Small 24B Instruct 2501 Quantized.w8a8?
11 devices with unified memory can run Mistral Small 24B Instruct 2501 Quantized.w8a8 at BF16 (47.9 GB), including Mac Mini M4 Pro (48 GB), Mac Pro M2 Ultra (192 GB), Mac Studio M2 Ultra (192 GB), Mac Studio M4 Max (128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.