Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

Mistral releases Codestral Mamba for faster, longer code generation

Jul 17, 2024 - news.bensbites.com
French AI startup Mistral has launched two new large language models (LLMs), Codestral Mamba 7B and Mathstral 7B. Codestral Mamba, based on the Mamba architecture, offers faster response times and longer context, outperforming rival open source models in benchmarking tests. It is designed for code productivity and will be free to use on Mistral’s la Plateforme API. Mathstral 7B, developed with Project Numina, is designed for math-related reasoning and scientific discovery, and has outperformed all other models designed for math reasoning.

Both models can be modified and deployed from their GitHub repositories and through HuggingFace, and are available under an open source Apache 2.0 license. Mistral, which recently raised $640 million in series B funding, bringing its valuation close to $6 billion, is competing against other AI developers like OpenAI and Anthropic. The company has also received investments from tech giants like Microsoft and IBM.

Key takeaways:

  • The French AI startup Mistral has launched two new large language models (LLMs), a math-based model and a code generating model, based on the new Mamba architecture.
  • The code generating model, Codestral Mamba 7B, offers a fast response time even with longer input texts and outperformed rival open source models in benchmarking tests.
  • Mistral's second model, Mathstral 7B, is designed specifically for math-related reasoning and scientific discovery, and outperformed every model designed for math reasoning.
  • Mistral recently raised $640 million in series B funding, bringing its valuation close to $6 billion, and received investments from tech giants like Microsoft and IBM.
View Full Article

Comments (0)

Be the first to comment!