Devstral 2 2512: price, context window and capabilities
Prices updated 8 Nov 2026
Devstral 2 2512 is a model from Mistral AI (mistralai/devstral-2512). It supports tool use. It costs $0.400 per million input tokens and $2.00 per million output tokens, with a 262K-token context window.
What it costs in practice
Using the listed prices: a chat turn of 2,000 input and 500 output tokens costs about $0.0018, so 10,000 such turns cost about $18.00. Summarising a 100,000-token document into 2,000 tokens costs about $0.044. A long agent run that reads 1 million tokens and writes 50,000 costs about $0.500.
Devstral 2 2512 compared with its closest rivals
| Model | Input / 1M | Output / 1M | Context | Chat turn |
|---|---|---|---|---|
| Devstral 2 2512 Mistral AI | $0.400 | $2.00 | 262K | $0.0018 |
| Llama 4 Maverick Meta | $0.350 | $1.00 | 524K | $0.0012 |
| Llama 3.3 70B Instruct Meta | $0.590 | $0.790 | 131K | $0.0016 |
| Llama 4 Scout Meta | $0.100 | $0.300 | 1.3M | $0.0003 |
Rivals are the models from other providers with the same tool and reasoning support whose output price is closest to Devstral 2 2512's.
How it compares
Among the 40 models we track, Devstral 2 2512 ranks #16 by input price and #15 by output price (1 is cheapest), and #30 by context window (1 is largest). Models with the same tool and reasoning support that cost less per output token: Llama 4 Scout ($0.300), Llama 3.3 70B Instruct ($0.790), Llama 4 Maverick ($1.00). Models with a larger context window at the same or lower output price: Llama 4 Scout (1.3M), DeepSeek V4.1 Flash 1017 (1.0M), DeepSeek V4.1 Flash (1.0M), DeepSeek V4 Flash 0731 (1.0M).
More from Mistral AI
All Mistral AI models on one page: Mistral AI API pricing.
- Mistral Large 4: $0.700 in / $2.10 out, 1.0M context
- Mistral Medium 3.5: $1.50 in / $7.50 out, 262K context
- Mistral Small 4: $0.150 in / $0.600 out, 262K context
Prices and limits come from OpenRouter's public model catalogue, last synced 8 Nov 2026. Provider pricing can differ from OpenRouter's and changes often; confirm on the provider's own page before you commit. See the guide to choosing a model or the full comparison.
Feed, daily deep-dive and bytes — readable offline, with push alerts for the topics you follow.