How much VRAM does Devstral Small 1.1 need?

Devstral Small 1.1 (24B parameters) requires approximately 21.2 GB of memory with Q4_K_M quantization.

What is the best quantization for Devstral Small 1.1?

The recommended quantization for Devstral Small 1.1 is Q4_K_M, which balances quality and memory efficiency.

Can it run?

No, Devstral Small 1.1 requires more memory than RX 7600 XT 16GB provides.

FWon't run

Too heavy

Using Q4_K_M in Ollama

Capabilities:

Fit status

Too heavy

Decode

14.2 tok/s

TTFT

13669 ms

Safe context

12K

Memory

21.2 GB / 16.0 GB

Weights14.6 GB

KV Cache3.8 GB

Runtime1.2 GB

Headroom1.6 GB

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	F	Too heavy	11.4 tok/s	24680 ms	21K
Chat	F	Too heavy	11.4 tok/s	9255 ms	7K
Coding	F	Too heavy	14.2 tok/s	13669 ms	12K
RAG	F	Too heavy	11.4 tok/s	30851 ms	21K
Reasoning	F	Too heavy	11.4 tok/s	20053 ms	12K

How Devstral Small 1.1 (24B params) fits at each quantization level on RX 7600 XT 16GB (16.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_KBest for your GPU	2	9.4 GB	Low	C42
Q3_K_S	3	11.8 GB	Low	C44

Upgrade options

MacBook Pro M4 32GBBest value

C5.9 tok/s decode

~$799 MSRP

RX 7900 XTX 24GBBudget pick

C47.2 tok/s decode

~$999 MSRP

RTX 5090 32GBBiggest leap