Mistral: Devstral 2 2512

Text input Text output Free Option
Author's Description

Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer model supporting a 256K context window. Devstral 2 supports exploring codebases and orchestrating changes across multiple files while maintaining architecture-level context. It tracks framework dependencies, detects failures, and retries with corrections—solving challenges like bug fixing and modernizing legacy systems. The model can be fine-tuned to prioritize specific languages or optimize for large enterprise codebases. It is available under a modified MIT license.

Key Specifications
Cost
$$$
Context
262K
Parameters
123B (Rumoured)
Released
Dec 09, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Max Tokens Temperature Top P Response Format Frequency Penalty Presence Penalty Tools Structured Outputs Seed Tool Choice
Features

This model supports the following features:

Tools Response Format Structured Outputs
Performance Summary

Mistral's Devstral 2 2512 model, released December 9, 2025, is a 123B-parameter dense transformer specializing in agentic coding with a 256K context window. It consistently ranks among the fastest models and offers highly competitive pricing, placing in the Infinityth percentile for both speed and cost across five benchmarks. The model demonstrates exceptional reliability with a 99% success rate, indicating minimal technical failures. In terms of performance, Devstral 2 exhibits a significant strength in its core domain, achieving 88.0% accuracy in the Coding benchmark, placing it in the 63rd percentile. It also performs well in General Knowledge with 98.5% accuracy (60th percentile) and Instruction Following at 62.0% accuracy (67th percentile). A notable weakness is its 0.0% accuracy in the Hallucinations (Baseline) test, suggesting it does not appropriately acknowledge uncertainty when presented with fictional concepts. Its Ethics performance is moderate at 96.0% accuracy (28th percentile), but with a very long duration. Overall, Devstral 2 excels in coding-related tasks and general knowledge, while its ability to identify and defer on fictional information requires improvement.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.15
Completion $0.6

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Chutes
Chutes | mistralai/devstral-2512 262K $0.15 / 1M tokens $0.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by mistralai