Glyd
GitHubGet started

Mistral-Small-3.2-24B-Instruct-2506

  • Mistral AI
  • 24B parameters
  • Dense
  • Apache-2.0
  • Released in bf16
  • Measured Sep 26, 2026

mistralai/Mistral-Small-3.2-24B-Instruct-2506 on Hugging Face

Raw logsRun it

Mistral-Small-3.2-24B-Instruct-2506 with Glyd

bf16GlydChange, %
Weights in GPU memory, GBEvery weight restored bit for bitlog48.0232.18−33log
Smallest single GPU, GBWeights, an 8K-token KV cache and 1.5 GB for the runtime4848
H100 80 GB GPUs it takes11
Its matrices, GBPacked and unpacked, every one bit for bitlog45.330.3−33log

Which single GPU it fits

Worked out from the measured weights.

MemoryGPUsbf16Glyd
16 GBRTX 4080, RTX 5080NoNo
24 GBRTX 4090, RTX 3090, A10NoNo
32 GBRTX 5090NoNo
48 GBRTX A6000, L40S, RTX 6000 AdaFitsFits
80 GBH100, A100 80 GBFitsFits
96 GBRTX PRO 6000, GH200FitsFits
141 GBH200FitsFits

Run it

pip install "glyd[gpu]"
import glyd
model = glyd.from_pretrained("mistralai/Mistral-Small-3.2-24B-Instruct-2506")

Linux and an NVIDIA GPU, Ampere or later: the weights are packed on the GPU as the model loads, every one bit for bit (getting started). vLLM serves Glyd's weights too, for the models checked there (serving with vLLM).

Related models

Also measured

Twenty-one more open models measured, with quality wherever bf16 fits one GPU.