Mistral
Devstral 2 123B Instruct (123B parameters) requires approximately 81.9 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 95 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run Devstral 2 123B Instruct on your machine.
Run
lms load Devstral-2-123B-Instruct-2512 && lms server startQuick specs
About this model
Quick picks
Best hardware
Run this model
Quantization options
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 48.0 GB | Low | — |
Q3_K_S | 3 | 60.3 GB | Low | — |
NVFP4 | 4 | 68.9 GB | Medium | — |
Q4_K_M | 4 | 75.0 GB | Medium | — |
Q5_K_M | 5 | 88.6 GB | High | — |
Q6_K | 6 | 100.9 GB | High | — |
Q8_0 | 8 | 131.6 GB | Very High | — |
F16 | 16 | 252.2 GB | Maximum | — |
Quality benchmarks
Coding
Source: official · 2025-12-18
Hardware compatibility
Computing compatibility...
Memory breakdown
Frequently asked questions
Devstral 2 123B Instruct (123B parameters) requires approximately 81.9 GB of VRAM with Q4_K_M quantization. Lower quantizations like Q4_K_M use less memory but may reduce quality.
Yes, Mac Studio M3 Ultra 256GB can run Devstral 2 123B Instruct with a compatibility score of 90/100. It provides 256 GB of memory and achieves approximately 8.1 tokens per second.
The recommended quantization for Devstral 2 123B Instruct is Q4_K_M, which offers the best balance between model quality and memory efficiency. Higher quantizations preserve more quality but require more VRAM.
The top recommended hardware for Devstral 2 123B Instruct: AMD Instinct MI300A 128GB (score: 98/100), NVIDIA H200 141GB (score: 98/100), NVIDIA H200 PCIe 141GB (score: 98/100). These provide the best combination of memory, bandwidth, and compute for running this model locally.
Yes, Devstral 2 123B Instruct is well-suited for coding as well as reasoning, agentic, vision. It was designed with these use cases in mind.
See also
