Voxtral Small 2507
Mistral AI
At a glance
- Price per 1M tokens (input)
- $0.10
- Price per 1M tokens (output)
- $0.40
- Context
- 32,768 tokens
- Free access
- yes (see below)
Refreshed daily; data verified August 30, 2026. Published August 30, 2026.
Voxtral Small 2507 is a compact language model from Mistral AI designed for developers and enterprises that need reliable function calling within their applications. It excels at tasks like API orchestration, database queries, and multi-step workflows where the model must decide when and how to invoke external tools. Mistral's approach emphasizes efficiency and transparency, offering a model that balances performance with operational simplicity. This model suits teams that want a lightweight yet capable assistant for structured automation without the overhead of larger systems.
Specifications & pricing
| Input (per 1M tokens) | $0.10 |
|---|---|
| Output (per 1M tokens) | $0.40 |
| Context window | 32,768 tokens |
| Max output | 32,768 tokens |
| Capabilities | tool calling |
LiteLLM community dataset (MIT), verified August 30, 2026. Official Mistral AI pricing.
What Voxtral Small 2507 would cost on your workload — run it through the cost calculator →
Where to try Voxtral Small 2507 for free
- Mistral AI offers a free chat — Le Chat (free plan). A vendor's free chat may run a different model from the same family — the exact model is not guaranteed.
Frequently asked questions
What is Voxtral Small 2507 best used for?+
It is optimized for tool calling, meaning it can parse user requests and trigger predefined functions or APIs. Typical use cases include virtual assistants, workflow automation, and any system where the model must interact with external software.
How do I integrate it with my existing stack?+
You can access it through Mistral's platform or API, and it supports standard function-calling protocols. The model accepts a list of available tools in the prompt and returns structured calls, which you can then execute in your backend.
How does it differ from other Mistral models?+
Unlike larger general-purpose models, Voxtral Small 2507 focuses on precision in tool invocation, with a leaner architecture that reduces latency and cost. It is not designed for open-ended creative writing but for deterministic, action-oriented tasks.
What are its limitations?+
It may struggle with complex reasoning or nuanced language understanding when not related to tool usage. Also, it requires careful definition of available functions to avoid ambiguity, and it may not handle very long conversational contexts without truncation.
Compare with others
Org chart: how to move your company onto AI
A practical map: which company roles and processes AI agents can take over, where to start, and in what order to roll it out.