Text model · Mistral
Devstral 2 123B: RAM and VRAM requirements
Devstral 2 123B needs about 73.3 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~69.75 GB to download; KV cache and overhead add the rest), or about 127.2 GB at Q8_0. The lightest hardware that runs it is Apple M4 Max (128GB).
Mistral family · 123B params · released Dec 2025 · 353.3K Ollama pulls.
Shopping for hardware? See what runs Devstral 2 123B →
Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.
Will it run on your device?
Devstral 2 123B runs on 4 of 43 tracked devices at Q4_K_M.
Memory breakdown
How context length changes it
Longer context grows the KV cache quickly: Devstral 2 123B needs ~73.3 GB at 4k but ~155.8 GB at 128k, which can push it past a device that fits it at a short context.
Speed drops too: every token re-reads the KV cache, so at 128k context Devstral 2 123B generates at roughly ~45% of its short-context speed. How this is estimated.
Quantization sizes
| Quantization | Size on disk |
|---|---|
| Q2_K | 51.5 GB est |
| Q3_K_M | 60.1 GB est |
| Q4_K_M (default) | 69.75 GB |
| Q5_K_M | 87.6 GB est |
| Q6_K | 100.9 GB est |
| Q8_0 | 123.73 GB |
| FP16 | 250.06 GB |
Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.
Run it
ollama run devstral-2:123b llama-cli -hf unsloth/Devstral-2-123B-Instruct-2512-GGUF:Q4_K_M lms get unsloth/Devstral-2-123B-Instruct-2512-GGUF Which devices can run Devstral 2 123B?
NVIDIA GPUs
- Nvidia GeForce RTX 2060 (6GB) No
- Nvidia GeForce RTX 3060 Ti (8GB) No
- Nvidia GeForce GTX 1070 (8GB) No
- Nvidia GeForce RTX 3060 (12GB) No
- Nvidia GeForce RTX 4070 (12GB) No
- Nvidia GeForce RTX 4060 Ti (16GB) No
- Nvidia GeForce RTX 4080 (16GB) No
- Nvidia GeForce RTX 4090 (24GB) No
- Nvidia GeForce RTX 3090 (24GB) No
- Nvidia GeForce RTX 5090 (32GB) No
Apple Silicon Macs
- Apple M1 (8GB) No
- Apple M2 (16GB) No
- Apple M4 (16GB) No
- Apple M5 (16GB) No
- Apple M3 Pro (18GB) No
- Apple M4 (24GB) No
- Apple M4 Pro (24GB) No
- Apple M5 (32GB) No
- Apple M4 Pro (48GB) No
- Apple M5 Pro (48GB) No
- Apple M4 Max (64GB) No
- Apple M4 Max (128GB) Yes
- Apple M5 Max (128GB) Yes
- Apple M3 Ultra (256GB) Yes
RAM-only laptops
iPhone & iPad
Android
Same job, different size
Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.
Similar models
Head-to-head
- Devstral 2 123B vs Ministral 3 3B
- Devstral 2 123B vs Mistral 7B
- Devstral 2 123B vs Ministral 3 8B
- Devstral 2 123B vs Mistral Nemo 12B
- Devstral 2 123B vs Ministral 3 14B
- Devstral 2 123B vs Mistral Small 3 24B
- Devstral 2 123B vs Mistral Small 3.1 24B
- Devstral 2 123B vs Magistral Small
- Devstral 2 123B vs Devstral Small
- Devstral 2 123B vs Devstral Small 2 24B
- Devstral 2 123B vs Mixtral 8x7B
FAQ
How much VRAM or RAM does Devstral 2 123B need?
At Q4_K_M, Devstral 2 123B needs about 73.3 GB (weights ~69.75 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~127.2 GB.
What is the Q4_K_M GGUF file size of Devstral 2 123B?
The Q4_K_M GGUF file is about 69.75 GB to download, and the Q8_0 GGUF is about 123.73 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~73.3 GB of memory at Q4_K_M.
Can Devstral 2 123B run on a laptop?
Devstral 2 123B is large; you need a high-memory Mac or multi-GPU setup at Q4_K_M.
Can I use Devstral 2 123B commercially?
Conditionally. Modified MIT: free below a stated revenue threshold, extra terms above it (see the LICENSE file).
Understand the numbers
Short guides to the ideas behind Devstral 2 123B's memory and quant figures.
Mistral's large agentic coding model: dense 123B, 256K context. Q4_K_M is a 74.9GB two-file split from unsloth, so it needs a 96GB-class machine even quantized. Modified MIT license with a revenue threshold.
Sources
Last validated 2026-09-07. Memory figures are estimates. See methodology.