Mistral Small 24B Instruct 2501
Neural Network
Mistral Small 24B Instruct 2501 is a 24B parameter language model by Mistral AI, optimized for low latency response.
Max answer length
(in tokens)
Context size
(in tokens)
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Mistral Small 24B Instruct 2501?
Mistral Small 24B Instruct 2501 is a 24B parameter language model from Mistral AI, built around one idea: fast responses for everyday tasks. The developer releases it under the Apache 2.0 license in two variants—pretrained and instruction-tuned. The Instruct version is designed for direct user prompts. Its compact size ensures low latency where reaction speed matters more than deep reasoning. In BotHub, the model is accessible from Russia without a VPN or foreign card: you pay in rubles only for tokens used, and they do not expire. Over 250 models are available in the same window, making it easy to switch and compare answers for the same prompt. The unified BotHub API allows you to connect Mistral Small to your product and switch models later without rewriting the integration. It is useful for responsive support assistants, bulk text processing and labeling, data extraction from documents, drafting emails and standard materials, and as a fast auxiliary layer alongside heavier models.Frequently asked questions about Mistral Small 24B Instruct 2501
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
This is a Mistral text model with a 32,000 token context. It is suitable for writing and editing text, coding, answering questions about uploaded documents and images, and function calling scenarios like chatbots, assistants, and internal company services.
The output limit is 16,384 tokens per response, which is tens of thousands of characters: enough for a long article, detailed analysis, or a large code snippet. The total dialogue context is 32,000 tokens, including both the request and the response.
A separate reasoning mode with a visible chain of thought is not noted in our data. However, you can ask the model in the prompt to break down the task step-by-step, check the solution process, and only then provide the output—this significantly helps with complex tasks.
Yes, the model supports function calling and structured JSON output. This is convenient for integrations: the model decides which tool to call and returns the answer in the specified schema. In BotHub, everything works via a unified OpenAI-compatible API.
In addition to text, images and documents are accepted as input: you can show a screenshot, diagram, or send a file and ask questions about the content. For audio, transcription, or video processing, switch to a specialized model—in BotHub, this is done in the same window.
Mistral claims its instruct models are multilingual, and Russian is among the supported languages, so the model handles standard work tasks well. It is easier to check if the style of the answers suits you by using your own prompts—access in BotHub works without a VPN or foreign card.