Context Window
—
Input Price/1M
—
Output Price/1M
—
Parameters
—
Speed
60 tok/s
Latency (TTFT)
672ms
Devstral 2 results on the main AI model evaluation benchmarks. Higher scores indicate better performance.
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| Terminal-Bench Hard | 19.0 | 100.0 | — |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| LiveBench Coding | 66.8 | 100.0 | Contamination-free benchmark with objective ground-truth answers |
| LiveCodeBench | 45.0 | 100.0 | Artificial Analysis official API |
| SciCode | 33.0 | 100.0 | — |
| AA Coding Index | 31.3 | 100.0 | Artificial Analysis official API |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| LiveBench Data Analysis | 39.1 | 100.0 | Contamination-free benchmark with objective ground-truth answers |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| MMLU-Pro | 76.0 | 100.0 | — |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| LiveBench Language | 45.7 | 100.0 | Contamination-free benchmark with objective ground-truth answers |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| AA-LCR | 30.0 | 100.0 | — |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| LiveBench Math | 52.5 | 100.0 | Contamination-free benchmark with objective ground-truth answers |
| AIME 2025 | 37.0 | 100.0 | Artificial Analysis official API |
| AA Math Index | 36.7 | 100.0 | Artificial Analysis official API |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| LiveBench Global | 41.2 | 100.0 | Contamination-free benchmark with objective ground-truth answers |
| AA Intelligence Index | 19.2 | 100.0 | Artificial Analysis official API |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| MMLU Pro | 76.2 | 100.0 | Artificial Analysis official API |
| GPQA Diamond | 59.0 | 100.0 | Artificial Analysis official API |
| IFBench | 38.0 | 100.0 | — |
| LiveBench Reasoning | 27.7 | 100.0 | Contamination-free benchmark with objective ground-truth answers |
| HLE | 4.0 | 100.0 | — |
| Benchmark | Score | Maximum | Methodology |
|---|---|---|---|
| Tau²-Bench | 25.0 | 100.0 | — |
Devstral 2 is an AI model developed by Mistral, classified as a text model. It focuses on text processing and natural language generation. As a proprietary model, it is available via Mistral's cloud API.
Devstral 2 does not have public per-token pricing available at this time. Some models offer access via enterprise plans or research programs. Check Mistral's official website for up-to-date availability and pricing.
Devstral 2 was evaluated on 20 different benchmarks, covering categories like Agentic, Coding, Data Analysis, Knowledge, Language, Long Context, Math, overall, Reasoning, Tool Use. Results show moderate performance across available evaluations.
It's important to note that benchmarks measure specific aspects and don't capture the full user experience. Factors like instruction adherence, behavior in long conversations, and real-world task quality vary significantly between models and aren't always reflected in standard scores.
Devstral 2 specializes in text, offering advanced capabilities for creating and processing text content.
In the 2026 AI model ecosystem, Devstral 2 competes directly with similarly capable models. Mistral competes in this segment against OpenAI, Anthropic, Google, and Meta. The choice between models depends on the specific use case, budget, latency requirements, and need for features like multimodality and tool calling.
For a detailed side-by-side comparison, use our comparison tool or check the overall model ranking.
Devstral 2 is an AI model developed by Mistral. It is a text model.
Devstral 2 does not have public per-token pricing available at this time. Check Mistral's official website for up-to-date information.
In available benchmarks, Devstral 2 scored: Terminal-Bench Hard: 19/100, LiveBench Coding: 66.79/100, LiveCodeBench: 45/100. See the full table above for a detailed comparison.
No, Devstral 2 is a proprietary model from Mistral. It is available via cloud API. For open source alternatives, check our open source model ranking.
Devstral 2 excels at general-purpose language tasks.
Last updated: July 17, 2026 • View methodology →