Capacité · 2026-06-29
Modèles d'IA avec support du Tool calling
Comparatif des modèles supportant le tool calling / function calling pour agents et flux automatisés.
Qu'est-ce que c'est ?
- Le tool calling (aussi appelé function calling) permet au LLM d'émettre une requête JSON structurée pour invoquer des fonctions externes — recherche, exécution de code, requêtes base de données, etc.
- Le modèle renvoie le nom de la fonction et ses arguments en JSON ; votre runtime les exécute et retourne le résultat sous forme de tool message.
Pourquoi c'est important
- Sans tool calling, les agents doivent parser du texte libre avec des expressions régulières fragiles.
- Le tool calling est la clé pour que RAG, boucles ReAct et assistants multi-étapes fonctionnent de manière fiable en production.
745 modèles avec cette capacité
Top 60 sur 745 affichés. Utilisez le répertoire complet pour filtrer davantage.
Frequently asked questions
How many AI models support tool calling?
745 canonical models in our database currently support tool calling. The list is regenerated on every data refresh, so it always reflects the latest releases tracked in our catalogue.
What is the cheapest model with tool calling?
Voxtral Small 24B 2507 from Mistral is currently the lowest-priced option, at $0.002 per 1M input tokens and $0.002 per 1M output tokens. The full table above is sorted price-ascending.
Which model with tool calling has the largest context window?
Qwen Long (Alibaba (Qwen)) leads on context at 10M tokens. This may matter if you also need long-document understanding alongside tool calling.
Which models are available on the most providers?
Production-readiness usually correlates with how many independent providers host the same weights. The top three by provider count are: Kimi K2.6 (49), Kimi K2.5 (48), GLM-5.1 (47).
How is tool calling different from a regular LLM?
Tool calling lets the model emit a structured JSON request to invoke an external function (search, code execution, DB query) instead of replying with prose. Without it, agents must parse freeform text — fragile and slow.
How often is this list updated?
Daily. Our data pipeline syncs once a day, regenerates the canonical model list, and rebuilds these pages so newly released models appear within 24 hours.
Explore more
Top models with this capability
- Voxtral Small 24B 2507$0.00 in / $0.00 out
- Ling-2.6-flash$0.01 in / $0.03 out
- Meta-Llama-3.1-8B-Instruct$0.02 in / $0.03 out
- Meta Llama 3.1 8B Instruct Turbo$0.02 in / $0.03 out
- Mistral Nemo$0.02 in / $0.04 out
Other capabilities
Best-of lists you might also want
Pricing comparisons
Vendors in this list
Dernière mise à jour :
Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.
Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.