unsloth / mistral family / image-text-to-text

mistral-small-3.2-24b-instruct-2506

mistral/mistral-small-3.2-24b-instruct-2506

GGUF Provisional scores Updated 26 Aug 2025

Run locally Smallest indexed GGUF / needs Ollama

ollama run hf.co/mistral/mistral-small-3.2-24b-instruct-2506/Mistral-Small-3.2-24B-Instruct-2506-UD-IQ1_S.gguf

Benchmarks

Where it scores.

Methodology →

General Provisional

35.8

Coding Provisional

46.1

Math Provisional

46.1

Reasoning Provisional

46.1

Normalized 0–100 across indexed models.

Runnable variants

Pick a size.

26 indexed

Precision	Format	Size	Params	File
IQ1_S / smallest	GGUF	5.6 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-IQ1_S.gguf
IQ1_M	GGUF	6.0 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-IQ1_M.gguf
IQ2_XXS	GGUF	6.8 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-IQ2_XXS.gguf
IQ2_M	GGUF	8.2 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-IQ2_M.gguf
Q2_K	GGUF	8.9 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q2_K.gguf
Q2_K_L	GGUF	9.0 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q2_K_L.gguf
Q2_K_XL	GGUF	9.3 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-Q2_K_XL.gguf
IQ3_XXS	GGUF	9.4 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-IQ3_XXS.gguf
Q3_K_S	GGUF	10.4 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q3_K_S.gguf
Q3_K_M	GGUF	11.5 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q3_K_M.gguf
Q3_K_XL	GGUF	11.9 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-Q3_K_XL.gguf
IQ4_XS	GGUF	12.8 GB	—	Mistral-Small-3.2-24B-Instruct-2506-IQ4_XS.gguf
IQ4_NL	GGUF	13.5 GB	—	Mistral-Small-3.2-24B-Instruct-2506-IQ4_NL.gguf
Q4_0	GGUF	13.5 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q4_0.gguf
Q4_K_S	GGUF	13.5 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q4_K_S.gguf
Q4_K_M	GGUF	14.3 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q4_K_M.gguf
Q4_K_XL	GGUF	14.5 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-Q4_K_XL.gguf
Q4_1	GGUF	14.9 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q4_1.gguf
Q5_K_S	GGUF	16.3 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q5_K_S.gguf
Q5_K_M	GGUF	16.8 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q5_K_M.gguf
Q5_K_XL	GGUF	16.8 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-Q5_K_XL.gguf
Q6_K	GGUF	19.3 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q6_K.gguf
Q6_K_XL	GGUF	20.8 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-Q6_K_XL.gguf
Q8_0	GGUF	25.1 GB	—	Mistral-Small-3.2-24B-Instruct-2506-Q8_0.gguf
Q8_K_XL	GGUF	29.0 GB	—	Mistral-Small-3.2-24B-Instruct-2506-UD-Q8_K_XL.gguf
BF16	GGUF	47.2 GB	—	Mistral-Small-3.2-24B-Instruct-2506-BF16.gguf

Sizes are file sizes, not peak runtime memory.

Raw benchmarks

The inputs.

3 records

Source	Task	Raw	Normalized
ZeroEval	GPQA	46.1	46.1
ZeroEval	MMMU	62.5	62.5
ZeroEval	SimpleQA	12.1	12.1

From the Hugging Face model card

About this model.

Read it on Hugging Face ↗