Question 1

Can I run Mistral Small 24B on a NVIDIA GeForce RTX 4090?

Accepted Answer

Yes — NVIDIA GeForce RTX 4090 runs Mistral Small 24B fully on GPU at the Q4_K_M quantization. Full GPU inference — 24 GB VRAM meets the 20 GB recommendation.

Question 2

What quantization of Mistral Small 24B should I use on a NVIDIA GeForce RTX 4090?

Accepted Answer

For 24 GB VRAM on the NVIDIA GeForce RTX 4090, the Q4_K_M variant is the best fit. Estimated ~57.6 tokens/sec on the Q4_K_M quantization.

Question 3

How fast does Mistral Small 24B run on NVIDIA GeForce RTX 4090?

Accepted Answer

Roughly 57.6 tokens/sec for Q4_K_M. Real speed depends on context length, backend (Ollama, llama.cpp, LM Studio), and KV cache size.

Question 4

What if NVIDIA GeForce RTX 4090 is not enough for Mistral Small 24B?

Accepted Answer

Try a smaller quantization variant or run on CPU + RAM with the trade-off of slower generation.

Quantization	File Size	Min VRAM	Rec VRAM	Context	Verdict	Estimated tok/s
Q4_K_MBest fit	14 GB	16 GB	20 GB	8K / 32K	Full GPU	~57.6
Q8_0	25 GB	27 GB	32 GB	8K / 32K	Hybrid CPU+GPU	~12

Can I Run Mistral Small 24B on NVIDIA GeForce RTX 4090?

Share this matchup

Every Mistral Small 24B quantization on NVIDIA GeForce RTX 4090