MI

Devstral Small 2505

MistralGGUF by eaddarioUpdated 2025-07-17
Download on Hugging Face
Parameters23.6B
Context128K
Layers—
Architecturellama
Smallest quant9.0 GB
Quick answer

Devstral Small 2505 needs about 14.5 GB of memory at Q4_K_M (a 12.5 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 13 quants, from 9.0 GB (Q3_K_S) to 47.2 GB (F16).

Devstral Small 2505 – agentic coding LLM with 128k context

Devstral Small 2505 is a 23.6 billion‑parameter model fine‑tuned for software engineering tasks. It supports a 128k token context window and is text‑only, making it suitable for local deployment on commodity GPUs. The model is released under Apache 2.0.

Good at

  • Agentic coding and tool use
  • Large context handling
  • Local inference on single RTX 4090

Limits

  • No vision support
  • Requires 128k context for full capability
  • Higher quantizations need more VRAM

Best for

  • Software engineering agents
  • Codebase exploration
  • Local LLM deployment

Who it’s for. Developers and teams needing an open‑source, agentic coding model that can run on a single RTX 4090 or a Mac with 32 GB RAM. Enterprises may use it as a base for custom domain extensions.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S9.0 GB 10.6 GB
Too big Get file →
IQ3_S9.6 GB 11.2 GB
Too big Get file →
Q3_K_M10.0 GB 11.7 GB
Too big Get file →
IQ3_M10.3 GB 12.1 GB
Too big Get file →
Q3_K_L10.9 GB 12.7 GB
Too big Get file →
Q4_K_S11.6 GB 13.5 GB
Too big Get file →
IQ4_NL12.5 GB 14.5 GB
Too big Get file →
Q4_K_M12.5 GB 14.5 GB
Too big Get file →
Q5_K_S14.5 GB 16.7 GB
Too big Get file →
Q5_K_M15.3 GB 17.7 GB
Too big Get file →
Q6_K18.9 GB 21.6 GB
Too big Get file →
Q8_020.3 GB 23.3 GB
Too big Get file →
F1647.2 GB 53.3 GB
Too big Get file →

For 8 GB VRAM use Q3_K_S (file size 9.01 GB, ~10.6 GB memory). For 12 GB VRAM use Q4_K_S (11.63 GB, ~13.5 GB). For 16 GB VRAM use Q5_K_S (14.46 GB, ~16.7 GB). Higher quantizations increase memory needs accordingly.

What GPU do you need for Devstral Small 2505?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: the best quant that fits is Q3_K_M (10.0 GB file, about 11.7 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (12.5 GB file, about 14.5 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (20.3 GB file, about 23.3 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (20.3 GB file, about 23.3 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (20.3 GB file, about 23.3 GB in use).

Which Devstral Small 2505 quant should you choose?

Q4_K_M is the usual starting point for Devstral Small 2505: it balances file size and quality well. Here it is a 12.5 GB file.

The Q8_0 file is 20.3 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q3_K_S at 9.0 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters23.6B
Trained context131,072 tokens

Source

Quantized byeaddario
FamilyMistral
Licenseapache-2.0
Downloads (HF)1,103
Last updated2025-07-17

Frequently asked questions about Devstral Small 2505

How much VRAM does Devstral Small 2505 need?

About 14.5 GB at Q4_K_M with an 8K context window. That is the 12.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Devstral Small 2505 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Devstral Small 2505 GGUF?

It depends on the quant: Q3_K_S is 9.0 GB, IQ3_S is 9.6 GB, Q3_K_M is 10.0 GB, IQ3_M is 10.3 GB, Q3_K_L is 10.9 GB, Q4_K_S is 11.6 GB and 7 more.

What context length does Devstral Small 2505 support?

The GGUF header lists a trained context length of 131,072 tokens (128K).

Where can I download Devstral Small 2505 GGUF?

The files are on Hugging Face in the eaddario/Devstral-Small-2505-pruned-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Devstral Small 2505 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Devstral Small 2505 on 8 GB VRAM?

Yes, using the Q3_K_S quantization which requires about 10.6 GB of VRAM at 8k context.

Does Devstral Small 2505 support vision?

No, the vision encoder was removed during fine‑tuning; the model is text‑only.

What license is Devstral Small 2505 released under?

Apache 2.0, allowing commercial and non‑commercial use and modification.

Other GGUFs of Devstral-Small-2505

Similar models

Browse all →

More from eaddario

All by eaddario →

More Mistral models

All Mistral →

Other medium models

All medium models →