How much VRAM does Devstral Small 1.1 need?

Devstral Small 1.1 (24B parameters) requires approximately 23.0 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?

Yes, MacBook Pro M2 Max 32GB can run Devstral Small 1.1 with a C grade (Runs with offload). Expected decode speed: 17.8 tok/s.

CUsable

Runs with offload

Using Q4_K_M in Ollama

Capabilities:

Fit status

Runs with offload

Decode

17.8 tok/s

TTFT

10902 ms

Safe context

16K

Memory

23.0 GB / 23.0 GB

Weights14.6 GB

KV Cache3.8 GB

Runtime1.2 GB

Headroom3.5 GB

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	D	Very compromised	17.0 tok/s	16552 ms	28K
Chat	C	Tight fit	17.8 tok/s	5946 ms	9K
Coding	C	Runs with offload	17.8 tok/s	10902 ms	16K
RAG	D	Very compromised	17.0 tok/s	20690 ms	28K
Reasoning	C	Runs with offload	17.8 tok/s	12884 ms	16K

How Devstral Small 1.1 (24B params) fits at each quantization level on MacBook Pro M2 Max 32GB (23.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	9.4 GB	Low	D38
Q3_K_S	3	11.8 GB	Low	C40
NVFP4	4

Ollama

ollama run devstral-small-2507

HuggingFace

huggingface-cli download devstral-small-2507

Upgrade options

RX 7900 XTX 24GBBudget pick

C47.2 tok/s decode

~$999 MSRP

RTX 3090 24GBBest value

C44.8 tok/s decode

~$1,499 MSRP

RTX 5090 32GBBiggest leap

B82 tok/s decode

~$1,999 MSRP