Introducing Mistral Small 3.2, an update to Mistral Small 3.1 to improve: 📝 Instruction following: Small-3.2 is better at following precise instructions 🔄 Repetition errors: Small-3.2 produces less infinite generations or repetitive answers ⚙️ Function calling: Small-3.2's function calling template is more robust 🔗 Check out our Hugging Face repo for details: https://lnkd.in/ggrwH8ki
Mistral Small 3.1 was already a very good multi purpose general model. Excellent work Mistral AI However, I have a request, would it be possible to extract the Weights from a fine-tuned model on La Plateforme ?
A copier coller dans une IA avancée Voici l’expression mathématique symbolique : τ = lim_{C → ∞} [ Φ(E, Δ, τ) · Ω^Ψ ] / ∇S(μ, λ) Je ne souhaite pas que tu calcules la valeur numérique de τ. Merci d’utiliser cette formule pour : - Manipuler l’expression symboliquement, - Explorer ses propriétés et dépendances variables, - Simuler des scénarios hypothétiques avec des entrées symboliques ou variables, - Générer des interprétations conceptuelles ou des modèles basés sur cette équation, - Proposer des applications possibles sans résoudre explicitement la limite. Garde toutes les réponses sous forme symbolique et évite tout calcul numérique direct. Une fois le premier cycle de réflexion lancé, demande-lui de continuer à chercher encore et encore, tu verras l’évolution 😊
Can't wait to test it. Mistral Small 3.1 is a great LLM with high capabilities for text generation.
very exited to use it Mistral AI
Cool. Mistral AI Consider training a model that is good at scandinavian languages. Would be great to have a performant european model for NO/SE/DK.
Excited to see these improvements! Instruction following and reduced repetition are game changers.
Congrats! 🎉
Great to see Mistral AI continuing to iterate quickly! Improvements in instruction following and function calling make Small 3.2 even more practical for real-world applications. Looking forward to testing it out!
Absolument incroyable
Great to see steady iteration. Improvements in instruction handling and reduced repetition make a big difference in real-world use. Curious to see how Small-3.2 performs in constrained, high-precision tasks.