readygpu

Mistral Small 3.2 24B

24B parameters · 128k context · apache-2.0 · released 2025-06-19

chatcodingvision

What to run it on in the US

Fully in memory at 32k context, priced at retailers today.

All hardware that runs it

Everything we price, best value first. Two cards split the model by layers.

HardwareQuantFitsDecodePricetok/s per 1,000
Intel Arc Pro B60Q5_K_Mfits10 tok/s$65015.8
AMD Radeon RX 7900 XTXQ5_K_Mfits24 tok/s$1,53315.6
NVIDIA GeForce RTX 3090Q5_K_Mfits26 tok/s$1,66015.8
AMD Radeon AI PRO R9700Q8_0fits12 tok/s$1,7006.8
NVIDIA GeForce RTX 3090 TiQ5_K_Mfits28 tok/s$2,06013.7
NVIDIA GeForce RTX 4090Q5_K_Mfits28 tok/s$3,2808.6
NVIDIA GeForce RTX 5090Q8_0fits37 tok/s$7,2005.1
Mac Studio M4 Max 128GBQ8_0fits12 tok/s––
Mac mini M4 Pro 64GBQ8_0fits6 tok/s––
Framework Desktop Ryzen AI Max+ 395 128GBQ8_0fits5 tok/s––
Mac Studio M3 Ultra 512GBQ8_0fits18 tok/s––
NVIDIA DGX Spark 128GBQ8_0fits6 tok/s––
GMKtec EVO-X2 Ryzen AI Max+ 395 128GBQ8_0fits5 tok/s––
Mac Studio M3 Ultra 256GBQ8_0fits18 tok/s––
Mac mini M4 Pro 48GBQ8_0fits6 tok/s––
Mac Studio M4 Max 64GBQ8_0fits12 tok/s––
Mac Studio M3 Ultra 96GBQ8_0fits18 tok/s––
NVIDIA RTX PRO 6000 BlackwellQ8_0fits37 tok/s––
ASUS Ascent GX10 128GBQ8_0fits6 tok/s––
Framework Desktop Ryzen AI Max+ 395 64GBQ8_0fits5 tok/s––
Mac mini M4 32GBQ4_K_Mfits4 tok/s––
NVIDIA Tesla P40Q5_K_Mfits10 tok/s––
NVIDIA GeForce RTX 3060 12GBQ4_K_Moffload5 tok/s$36013.3
Intel Arc B580Q4_K_Moffload5 tok/s$37012.7
Intel Arc A770 16GBQ4_K_Moffload7 tok/s$40018.3
NVIDIA GeForce RTX 3080 10GBQ4_K_Moffload5 tok/s$4699.8
AMD Radeon RX 9060 XT 16GBQ4_K_Moffload6 tok/s$54310.9
AMD Radeon RX 7800 XTQ4_K_Moffload8 tok/s$59913.2
AMD Radeon RX 7600 XTQ4_K_Moffload6 tok/s$6958.1
NVIDIA GeForce RTX 4070 SuperQ4_K_Moffload5 tok/s$6997.4
NVIDIA GeForce RTX 4060 Ti 16GBQ4_K_Moffload6 tok/s$7707.9
NVIDIA GeForce RTX 5060 Ti 16GBQ4_K_Moffload8 tok/s$7729.7
NVIDIA GeForce RTX 5070Q4_K_Moffload5 tok/s$7767.0
AMD Radeon RX 7900 XTQ4_K_Moffload21 tok/s$80026.5
AMD Radeon RX 9070 XTQ4_K_Moffload8 tok/s$8159.7
NVIDIA GeForce RTX 4070 Ti SuperQ4_K_Moffload9 tok/s$1,0138.7
NVIDIA GeForce RTX 5070 TiQ4_K_Moffload10 tok/s$1,1448.4
NVIDIA GeForce RTX 4080 SuperQ4_K_Moffload9 tok/s$1,1997.5
NVIDIA GeForce RTX 5080Q4_K_Moffload10 tok/s$1,7385.6

1 other card or machine we price cannot hold it at this context.

Memory needed

Weights, KV cache and runtime buffers at each context length (KV cache at fp16).

QuantFile@8k@32k@64k@128k
Q4_K_Mgguf13.3 GB15.3 GB19.2 GB24.4 GB34.8 GB
mlx-4bitmlx12.3 GB14.3 GB18.2 GB23.4 GB33.8 GB
Q5_K_Mgguf15.6 GB17.6 GB21.5 GB26.7 GB37.1 GB
Q6_Kgguf18.0 GB20.0 GB23.9 GB29.1 GB39.5 GB
mlx-8bitmlx23.3 GB25.3 GB29.2 GB34.4 GB44.8 GB
Q8_0gguf23.3 GB25.3 GB29.2 GB34.4 GB44.8 GB

Run it

Copy a command for the runtime you use.

  • Q4_K_M
    • Ollamaollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_M
    • LM Studiolms get unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF@q4_k_m
    • llama.cppllama-server -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q4_K_M
  • mlx-4bit
    • LM Studiolms get mlx-community/Mistral-Small-3.2-24B-Instruct-2506-4bit
    • MLXmlx_lm.server --model mlx-community/Mistral-Small-3.2-24B-Instruct-2506-4bit
  • Q5_K_M
    • Ollamaollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q5_K_M
    • LM Studiolms get unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF@q5_k_m
    • llama.cppllama-server -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q5_K_M
  • Q6_K
    • Ollamaollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q6_K
    • LM Studiolms get unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF@q6_k
    • llama.cppllama-server -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q6_K
  • mlx-8bit
    • LM Studiolms get mlx-community/Mistral-Small-3.2-24B-Instruct-2506-8bit
    • MLXmlx_lm.server --model mlx-community/Mistral-Small-3.2-24B-Instruct-2506-8bit
  • Q8_0
    • Ollamaollama run hf.co/unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q8_0
    • LM Studiolms get unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF@q8_0
    • llama.cppllama-server -hf unsloth/Mistral-Small-3.2-24B-Instruct-2506-GGUF:Q8_0