Devstral — local AI model by Mistral AI
Written by Jakub Rusinowski · Last updated
Mistral AI's coding specialists, spanning both generations. The original Devstral Small 2505 posts 46.8% on SWE-Bench Verified; its December 2025 successor Devstral Small 2 lifts that to 68.0% at the same 24B, which is the local-agent sweet spot — about 15 GB at Q4_K_M on one consumer card. The 123B flagship reaches 71.6%. Built for agentic software engineering: multi-file editing, repo navigation, and test-driven development. Apache 2.0 throughout.
Variants
The smallest Devstral variant needs about 14 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| Devstral-2 123B → 123B | ~75.1 GB | 262,144 | ollama run devstral:123b | |
| Devstral 2 22B 22B | ~14.1 GB | 128,000 | ollama run devstral-2 | |
| Devstral Small 2505 24B → 24B | ~15.3 GB | 128,000 | ollama run devstral:24b | |
| Devstral Small 2 24B → 24B | ~15.3 GB | 262,144 | ollama run devstral-small-2:24b |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run Devstral locally
Install Ollama, then pull the tag.
ollama run devstral:123bPick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: Devstral-2 123B, Devstral 2 22B, Devstral Small 2505 24B, Devstral Small 2 24BRecommended GPU
The cheapest catalogued GPU that runs Devstral locally (min 14 GB VRAM) is the AMD Radeon RX 9060 XT 16GB (16 GB).
Can I run Devstral on my GPU?
- Devstral on AMD Radeon RX 7800 XT
- Devstral on AMD Radeon RX 7900 XT
- Devstral on AMD Radeon RX 7900 XTX
- Devstral on AMD Radeon RX 9060 XT 16GB
- Devstral on AMD Radeon RX 9060 XT 8GB
- Devstral on AMD Radeon RX 9070
- Devstral on AMD Radeon RX 9070 XT
- Devstral on AMD Ryzen AI Max+ 395
- Devstral on Apple M1
- Devstral on Apple M1 Pro
- Devstral on Apple M1 Ultra
- Devstral on Apple M2
- Devstral on Apple M2 Pro
- Devstral on Apple M3
- Devstral on Apple M3 Pro
- Devstral on Apple M4
- Devstral on Apple M4 Max
- Devstral on Apple M4 Pro
- Devstral on Apple M5
- Devstral on Apple M5 Max
- Devstral on Intel Arc B570
- Devstral on Intel Arc B580
- Devstral on NVIDIA A100 80GB (PCIe)
- Devstral on NVIDIA DGX Spark
- Devstral on NVIDIA GeForce RTX 3060 (12GB)
- Devstral on NVIDIA GeForce RTX 3070
- Devstral on NVIDIA GeForce RTX 3070 Ti
- Devstral on NVIDIA GeForce RTX 3080 (10GB)
- Devstral on NVIDIA GeForce RTX 3090
- Devstral on NVIDIA GeForce RTX 4060
- Devstral on NVIDIA GeForce RTX 4060 Ti 16GB
- Devstral on NVIDIA GeForce RTX 4070
- Devstral on NVIDIA GeForce RTX 4070 Super
- Devstral on NVIDIA GeForce RTX 4070 Ti
- Devstral on NVIDIA GeForce RTX 4070 Ti Super
- Devstral on NVIDIA GeForce RTX 4080
- Devstral on NVIDIA GeForce RTX 4080 Super
- Devstral on NVIDIA GeForce RTX 4090
- Devstral on NVIDIA GeForce RTX 5060
- Devstral on NVIDIA GeForce RTX 5060 Ti 16GB
- Devstral on NVIDIA GeForce RTX 5060 Ti 8GB
- Devstral on NVIDIA GeForce RTX 5070
- Devstral on NVIDIA GeForce RTX 5070 Ti
- Devstral on NVIDIA GeForce RTX 5080
- Devstral on NVIDIA GeForce RTX 5090
- Devstral on NVIDIA H100 80GB (PCIe)
- Devstral on NVIDIA RTX PRO 6000 Blackwell
Devstral — frequently asked questions
How much VRAM does Devstral need?
Devstral needs about 14 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Devstral-2 123B (75 GB, Q4_K_M); Devstral 2 22B (14 GB, Q4_K_M); Devstral Small 2505 24B (15 GB, Q4_K_M); Devstral Small 2 24B (15 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Devstral on an RTX 4090 (24 GB)?
Yes — Devstral runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for Devstral?
Q4_K_M is the best balance of quality and VRAM for Devstral in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Devstral with Ollama?
Install Ollama, then run: ollama run devstral:123b. This downloads Devstral and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.