tech
Mistral Small 3
Today we're introducing Mistral Small 3, a latency-optimized 24B-parameter model released under the Apache 2.0 license.
TL;DR
- Mistral Small 3 is a new 24B-parameter AI model released under the Apache 2.0 license.
- It offers low latency and performance comparable to larger models like Llama 3.3 70B.
- The model is designed for '80%' of generative AI tasks requiring robust language and instruction following with very low latency.
- It is suitable for local deployment, virtual assistants, low-latency function calling, fine-tuning for specific domains, and local inference on capable hardware.
- Mistral is committed to using the Apache 2.0 license for general-purpose models, with weights available for download and local deployment.