Quick Facts
| Developer | Mistral AI |
| License | Apache 2.0 -- permits commercial use subject to standard Apache terms; verify against the official model card |
| Typical Sizes | 24B -- cloud tag devstral-small-2:24b-cloud; local tag devstral-small-2:24b (~15 GB, 384K local context) also available on Ollama |
| Context | 256K tokens (cloud tag), native tool support |
| Modality | Text and image input |
| Best For | Agentic coding, codebase exploration, multi-file editing, test generation |
| Recommended Deployment | Terminal.Glass Cloud tier (low usage) for individuals and small teams; local deployment for steady daily use |
Why Choose This Model
Devstral Small 2 packs agentic coding capability into a compact 24B footprint -- tool use, file-tree exploration, multi-file editing, and test generation, plus image input for reading screenshots or diagrams alongside code. It's a practical entry point for individuals and small teams that want a capable coding agent without provisioning hardware for a larger model, and its low-usage cloud tier keeps per-token cost lower than frontier-scale coding models.
Terminal.Glass Deployment
Your interface, chat history, user accounts, and RAG index stay on your Terminal.Glass host.
Each request to devstral-small-2:24b-cloud
-- including code excerpts, images, and retrieved passages -- is sent to Ollama Cloud for inference
and streamed back. This tag is billed at Ollama's low-usage tier, though long 256K contexts and
image inputs can still accumulate spend under heavy agentic use. No local GPU is required for the
cloud tag; a small VM or mini-PC running Open WebUI and the Ollama client is sufficient.
Recommended Uses
- Codebase exploration and file-tree navigation for unfamiliar projects
- Multi-file edits, refactors, and tool-assisted debugging
- Test generation and code review preparation
- Reading screenshots, diagrams, or UI mockups alongside code (image input)
Not recommended for proprietary source code, client repositories, credentials, or sensitive screenshots -- all request content leaves your network for this tag. Smaller than full Devstral 2; validate complex changes with tests and review before merging. Generated multi-file edits and terminal-agent actions still require human review and rollback discipline.
Hardware Guidance
For the cloud tag, the Terminal.Glass host only needs to run Open WebUI, the Ollama client, and your
document index -- a small VM or mini-PC is enough, since inference happens on Ollama Cloud. Devstral
Small 2 is also available as a local weight release (devstral-small-2:24b,
~15 GB, 384K local context) for teams that want to run it directly on owned GPU hardware -- a
practical option once usage is steady and predictable. Confirm current tags at
the official Ollama listing.
Honest Guidance
- Regulated or confidential codebases: use the local
devstral-small-2:24btag on hardware you control instead of the cloud tag. - Steady daily team usage: local deployment on owned hardware may cost less over time than recurring cloud tokens and removes data-movement risk.
- Most demanding software-agent workloads: full Devstral 2 (123B) or another larger coding model may fit better when quality matters more than compact size.
- Version pinning or audit reproducibility: cloud tags can change server-side; the local tag offers a pinned, reproducible alternative.
- Unreviewed agent automation: multi-file edits, generated patches, and terminal-agent actions need human review, tests, and rollback plans regardless of model choice.
Learn More
Full technical reference and licensing detail for the Devstral family are documented on NoCloudGPT. Confirm current tags and availability at ollama.com/library/devstral-small-2.