FitMyLLM
Mistral AI/Dense

Mistral AIDevstral Small 22B

Devstral Small — Mistral's model for software engineering and agentic coding.

coding
23.57B
Parameters
128K
Context length
23
Benchmarks
10
Quantizations
50K
HF downloads
Architecture
Dense
Released
2025-05-07
Layers
40
KV Heads
8
Head Dim
128
Family
mistral

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
Q3_K_M4
14.1 GB
12.3 + 1.9 KV
low
Q3_K_L4.3
15.0 GB
13.2 + 1.9 KV
moderate
IQ4_XS4.46
15.5 GB
13.6 + 1.9 KV
moderate
Q4_K_S4.67
16.1 GB
14.2 + 1.9 KV
moderate
Q4_K_M4.89
16.8 GB
14.9 + 1.9 KV
good
Q5_K_S5.57
18.8 GB
16.9 + 1.9 KV
good
Q5_K_M5.7
19.2 GB
17.3 + 1.9 KV
good
Q6_K6.56
21.7 GB
19.8 + 1.9 KV
excellent
Q8_08.5
27.4 GB
25.5 + 1.9 KV
lossless
FP1616
49.5 GB
47.6 + 1.9 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
Ready to run this?Rent by the hour

Rent a GPU and run Devstral Small 22B now

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (23)

Run this model

Easiest way to get started·Beginners
Docs ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run hf.co/bartowski/Devstral-Small-2505-GGUF:Q4_K_M

Not in Ollama's library: Ollama pulls the GGUF from Hugging Face (bartowski/Devstral-Small-2505-GGUF). If Q4_K_M is not in that repo, pick a file listed there.

Setup guide
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

Apple M1 Pro (16GB)
16 GB VRAM • 200 GB/s
APPLE
$999
Apple M2 Pro (16GB)
16 GB VRAM • 200 GB/s
APPLE
$1299
Apple M4 (16GB)
16 GB VRAM • 120 GB/s
APPLE
$499
NVIDIA Tesla T4 16GB
16 GB VRAM • 320 GB/s
NVIDIA
$800
NVIDIA V100 PCIe 16GB
16 GB VRAM • 900 GB/s
NVIDIA
$2000
Apple M1 (16GB)
16 GB VRAM • 68.25 GB/s
APPLE
$699
Apple M2 (16GB)
16 GB VRAM • 100 GB/s
APPLE
$799
Apple M3 (16GB)
16 GB VRAM • 100 GB/s
APPLE
$799
NVIDIA Tesla P100 DGXS
16 GB VRAM • 732 GB/s
NVIDIA
NVIDIA Tesla P100 PCIe 16 GB
16 GB VRAM • 732 GB/s
NVIDIA
NVIDIA Tesla P100 SXM2
16 GB VRAM • 732 GB/s
NVIDIA
NVIDIA Tesla V100 PCIe 16 GB
16 GB VRAM • 897 GB/s
NVIDIA
NVIDIA Tesla V100 SXM2 16 GB
16 GB VRAM • 1130 GB/s
NVIDIA

Find the best GPU for Devstral Small 22B

Build Hardware for Devstral Small 22B

Devstral Small — Mistral's model for software engineering and agentic coding.

Spec sheet

Devstral Small 22B — 23.57B Dense.

Specifications
Parameters
23.57B
Architecture
Dense Transformer
Context Length
128K tokens
Capabilities
coding
Release Date
2025-05-07
Provider
Mistral AI
Family
mistral
VRAM requirements
QuantBPWVRAMQuality
Q3_K_M412.3 GB88%
Q3_K_L4.313.2 GB90%
IQ4_XS4.4613.6 GB92%
Q4_K_S4.6714.2 GB93%
Q4_K_M4.8914.9 GB94%
Q5_K_S5.5716.9 GB96%
Q5_K_M5.717.3 GB96%
Q6_K6.5619.8 GB97%
Q8_08.525.5 GB100%
FP161647.6 GB100%
Benchmark scores
HumanEval72.0
MMLU-PRO49.4
MATH35.6
IFEval65.7
BBH52.8
GPQA18.6
MUSR17.1
MBPP74.0
BigCodeBench36.1
GPQA Diamond53.2
LiveCodeBench34.8
AIME34.3
HLE3.4
AA Intelligence19.5
AA Coding20.7
AA Math34.3
IFBench31.2
Terminal-Bench Hard16.7
τ²-Bench23.4
SciCode28.8
AA Long Context24.0
Terminal-Bench 2.129.6
Terminal-Bench 40.0
Compatible GPUs
30 @ Q4_K_M