Happy Horse 1.0 is now on ModelsLab

Try Now
Skip to main content
Available now on ModelsLab · Language Model

Mistral Small (24B) Instruct 25.01Fast. Efficient. Production-Ready.

Try Mistral Small (24B) Instruct 25.01API Documentation

Built For Speed And Accuracy

Lightning-Fast Inference

150 Tokens Per Second

Delivers 81% MMLU accuracy with ultra-low latency for real-time applications.

Compact Architecture

24B Parameters, Full Power

Runs on single GPU or 32GB Mac. Competes with models three times its size.

Extended Context

32K Token Window

Process longer documents and conversations without losing context or quality.

Examples

See what Mistral Small (24B) Instruct 25.01 can create

Copy any prompt below and try it yourself in the playground.

Customer Support Agent

“You are a helpful customer support assistant. Answer questions about product features, pricing, and troubleshooting. Keep responses concise and professional. User question: How do I reset my password?”

Code Review

“Review this Python function for bugs and performance issues. Suggest improvements and explain your reasoning. Function: def calculate_total(items): total = 0; for item in items: total = total + item['price'] * item['quantity']; return total”

Content Summarization

“Summarize the following article in 3 bullet points, focusing on key takeaways. Article: [paste technical documentation or blog post]”

Multi-Language Translation

“Translate the following English text to Spanish, French, and German. Maintain formal tone. Text: The quarterly earnings report shows a 15% increase in revenue.”

For Developers

A few lines of code.
Fast inference. Three lines.

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

  • Serverless: scales to zero, scales to millions
  • Pay per token, no minimums
  • Python and JavaScript SDKs, plus REST API
API Documentation 
import requests


response = requests.post(
    "https://modelslab.com/api/v7/llm/chat/completions",
    json={
  "key": "YOUR_API_KEY",
  "prompt": "",
  "model_id": ""
}
)
print(response.json())

FAQ

Common questions about Mistral Small (24B) Instruct 25.01

Read the docs

Mistral Small 24B Instruct 2501 is a 24-billion parameter instruction-tuned language model optimized for low-latency text generation. It achieves performance in its category while maintaining exceptional speed and efficiency.

Despite having 24B parameters, it performs competitively with models three times its size on code, math, and general knowledge benchmarks. It delivers 150 tokens per second with 81% MMLU accuracy, making it ideal for production deployments.

Mistral Small 24B Instruct 2501 excels at conversational AI, code assistance, enterprise RAG, agentic systems, and multilingual tasks. It's perfect for applications requiring fast, accurate responses with minimal latency.

Yes. Once quantized, it fits on a single RTX 4090 GPU or a 32GB Mac, making it ideal for on-device deployment and handling sensitive data locally.

Mistral Small 24B Instruct 2501 supports a 32k token context window, allowing it to process longer documents and maintain conversation history effectively.

Yes, it's released under Apache 2.0 license, permitting both commercial and non-commercial usage and modification. Both pretrained and instruction-tuned checkpoints are available.

Ready to create?

Start generating with Mistral Small (24B) Instruct 25.01 on ModelsLab.

Try Mistral Small (24B) Instruct 25.01API Documentation