Discover Mistral: Devstral Small 1.1

Streamline coding tasks with Mistral: Devstral Small 1.1's capabilities.

Input: text Output: text Context: 128,000 tokens Release: 2025-07-10

Mistral: Devstral Small 1.1 is a language model developed by Mistral AI in collaboration with All Hands AI, released on July 10, 2025. This model, finetuned from Mistral Small 3.1, is tailored for software engineering workflows, offering a substantial 128k token context window. It supports text-to-text interactions, making it ideal for tasks like codebase exploration and multi-file edits. With its capability to run on a single 4090 GPU or Apple silicon machine, Devstral Small 1.1 is accessible for various development environments. It uses a Tekken tokenizer and supports Mistral-style function calling and XML outputs, ensuring flexibility in coding tasks. The model's performance on SWE-Bench Verified highlights its efficiency in handling complex software engineering challenges.

Use Cases

Here are a few ways teams apply Mistral: Devstral Small 1.1 in practice—from fast drafting to multimodal understanding. Adapt these ideas to your workflow.

Enhance codebase exploration

Perform multi-file code edits

Integrate into autonomous agents

Optimize software engineering workflows

Support with function calling tasks

Key Features

A quick look at the capabilities that make this model useful in real projects.

128k token context window

Supports Mistral-style function calling

XML output formats

Runs on single 4090 GPU

Tekken tokenizer with 131k vocabulary

OpenAI-compatible runtimes

Specs

Overview

Vendor
mistralai

Model ID
mistralai/devstral-small

Release
2025-07-10

Modalities & context

Input
text

Output
text

Context
128,000 tokens

Parameters & defaults

Supported parameters: frequency_penalty, max_tokens, min_p, presence_penalty, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_p

Defaults: temperature 0.3, top_p 0.95

Benchmark tests: Mistral: Devstral Small 1.1

We ran this model against a few representative prompts to show its range. Review the outputs below and be the judge.

Text

Prompt:

Write 150 words on how AI might positively upend work, leisure and creativity

AI has the potential to revolutionize work, leisure, and creativity in transformative ways. In the workplace, AI can automate repetitive tasks, freeing humans to focus on strategic, creative, and interpersonal roles. It can also enhance productivity through predictive analytics, personalized learning, and real-time decision-making support. In leisure, AI-powered personalization—such as tailored entertainment recommendations, virtual travel experiences, and adaptive gaming—can make downtime more engaging and fulfilling. For creativity, AI tools like generative design, music composition algorithms, and AI-assisted writing can democratize artistic expression, allowing more people to explore and refine their ideas. However, these benefits depend on ethical AI development, ensuring accessibility and preventing job displacement. By augmenting human capabilities rather than replacing them, AI could lead to a more balanced, innovative, and fulfilling future. The key lies in harnessing its potential responsibly, fostering collaboration between humans and machines to unlock new possibilities in every aspect of life.

Run this prompt on Upend.AI

Ready to try Mistral: Devstral Small 1.1?

Chat with Mistral: Devstral Small 1.1

up.end

/ˌəpˈend/

verb

To “upend” means to completely disrupt, overturn, or drastically change the established order or structure of something. It implies a significant shift or alteration that can potentially have far-reaching consequences. When something is upended, it is turned upside down or transformed in a way that challenges conventional norms or expectations. The term often carries a sense of innovation, transformation, and sometimes even a hint of upheaval, indicating that the changes are not just minor adjustments but rather a fundamental reimagining of the status quo.

Privacy Terms