Devstral 2 123B Instruct 2512
125B parameters · 256k context · Modified MIT License · released 2025-11-28
codingagenticchat
In short
Devstral 2 123B Instruct 2512 is an open model from Mistral AI with 125B parameters. At Q8_0, Devstral 2 123B Instruct 2512 needs about 127 GB of memory at 8k context and 169 GB at 128k, counting the weights, the KV cache and runtime buffers. Nothing we price in the US runs Devstral 2 123B Instruct 2512 fully in memory at 128k context at a usable speed; see the table for what comes closest.
Updated . Speeds are estimates; how they are worked out.
What to run it on in the US
Fully in memory at 128k context, priced at retailers today.
Something we price holds it fully in memory, but none reaches 15 tok/s, our bar for recommending it. The table below has the speeds.
All hardware that runs it
Everything we price, best value first. Two cards split the model by layers.
| Hardware | Quant | Fits | Decode | Price | tok/s per 1,000 |
|---|---|---|---|---|---|
| Mac Studio M3 Ultra 256GB | Q8_0 | fits | 3 tok/s | $8,499 | 0.4 |
| Mac Studio M3 Ultra 512GB | Q8_0 | fits | 3 tok/s | $23,000 | 0.1 |
38 other cards and machines we price cannot hold it at this context.
Memory needed
Weights, KV cache and runtime buffers at each context length (KV cache at fp16).
| Quant | File | @8k | @32k | @64k | @128k |
|---|---|---|---|---|---|
| Q8_0gguf | 123.7 GB | 127.2 GB | 135.6 GB | 146.8 GB | 169.2 GB |
Run it
Copy a command for the runtime you use.
- Q8_0
- Ollama
ollama run hf.co/ggml-org/Devstral-2-123B-Instruct-2512-GGUF:Q8_0 - LM Studio
lms get ggml-org/Devstral-2-123B-Instruct-2512-GGUF@q8_0 - llama.cpp
llama-server -hf ggml-org/Devstral-2-123B-Instruct-2512-GGUF:Q8_0
- Ollama
Notes
The weights are FP8, and the example vLLM/SGLang commands use tensor parallelism 8, which needs 8 GPUs. It needs mistral_common >= 1.8.6, and llama.cpp GGUF builds need the changes from llama.cpp PR #17945.
Questions
- How much VRAM does Devstral 2 123B Instruct 2512 need?
- At Q8_0, Devstral 2 123B Instruct 2512 needs about 127 GB of memory at 8k context and 169 GB at 128k, counting the weights, the KV cache and runtime buffers.
- Can I run Devstral 2 123B Instruct 2512 on a 16 GB GPU?
- No: it does not fit in 16 GB at 128k context, even with system RAM.
- Can I run Devstral 2 123B Instruct 2512 on a 24 GB GPU?
- No: it does not fit in 24 GB at 128k context, even with system RAM.
- How do I run Devstral 2 123B Instruct 2512 locally?
- With Ollama: ollama run hf.co/ggml-org/Devstral-2-123B-Instruct-2512-GGUF:Q8_0 (the Q8_0 build). LM Studio and llama.cpp commands are on this page.
- What licence is Devstral 2 123B Instruct 2512 released under?
- Modified MIT License.