Funktion · 2026-05-12

KI-Modelle mit Tool-Calling-Unterstützung

Vergleich von Modellen mit Tool Calling / Function Calling für Agenten und Automatisierung.

Was ist das?

Tool Calling (Function Calling) erlaubt es dem LLM, strukturierte JSON-Aufrufe zu externen Funktionen zu erzeugen — Suche, Codeausführung, Datenbankabfragen usw.
Das Modell liefert Funktionsnamen und Argumente als JSON; Ihre Runtime führt sie aus und gibt Tool-Messages zurück.

Warum wichtig

Ohne Tool Calling bleiben Agenten häufig auf fragiles Parsing von Freitext angewiesen.
Tool Calling macht RAG, ReAct-Schleifen und mehrstufige Assistenten in Produktion robust.

660 Modelle mit dieser Funktion

Modell	Anbieter	Eingabe / 1M	Ausgabe / 1M	Kontext	Hoster
Voxtral Small 24B 2507	Mistral	$0.002	$0.002	32K	3
Meta-Llama-3.1-8B-Instruct	Meta	$0.020	$0.030	128K	25
Llama 3.1 8B Turbo	Meta	$0.020	$0.030	131K	2
Mistral Nemo Instruct 2407	Mistral	$0.020	$0.040	128K	8
Hermes 4 14B	chutes	$0.014	$0.054	41K	1
Ministral 3B	azure-cognitive-services	$0.040	$0.040	128K	1
Ministral 3B (latest)	Mistral	$0.040	$0.040	128K	1
Llama 3.1 8B Instruct	Meta	$0.030	$0.050	131K	1
Ministral 3B	azure	$0.040	$0.040	128K	1
Llama-3.2-11B-Vision-Instruct	Meta	$0.049	$0.049	128K	8
L3 8B Stheno V3.2	novita-ai	$0.050	$0.050	8K	1
Llama 3.1 8B Instant	Meta	$0.050	$0.080	131K	2
Llama 3 8B	Meta	$0.050	$0.080	8K	1
Gemma 3 27B	Google	$0.027	$0.109	131K	14
Qwen2.5-Coder 32B Instruct	Alibaba (Qwen)	$0.027	$0.109	131K	7
Model Router	azure-cognitive-services	$0.140	Unknown	128K	1
Model Router	azure	$0.140	Unknown	128K	1
IBM: Granite 4.1 8B	kilo	$0.050	$0.100	131K	1
DeepSeek R1 Distill Llama 70B	Meta	$0.030	$0.130	131K	5
GPT OSS 20B	OpenAI	$0.030	$0.140	131K	23
Amazon: Nova Micro 1.0	kilo	$0.035	$0.140	128K	1
Nova Micro	amazon-bedrock	$0.035	$0.140	128K	1
Nova Micro	vercel	$0.035	$0.140	128K	1
Command R7B	Cohere	$0.037	$0.150	128K	2
Command R7B Arabic	Cohere	$0.037	$0.150	128K	1
Gemini 1.5 Flash-8B	Google	$0.037	$0.150	1M	1
Arcee AI: Trinity Mini	kilo	$0.045	$0.150	131K	1
Trinity Mini	clarifai	$0.045	$0.150	131K	1
GPT OSS 120B	OpenAI	$0.040	$0.160	131K	33
Qwen3 235B A22B Instruct 2507	Alibaba (Qwen)	$0.100	$0.100	262K	18
Qwen3-235B-A22B-Thinking-2507	Alibaba (Qwen)	$0.100	$0.100	262K	17
Qwen3 30B A3B Instruct 2507	Alibaba (Qwen)	$0.100	$0.100	262K	12
Qwen3-30B-A3B	Alibaba (Qwen)	$0.100	$0.100	128K	9
Qwen3 30B A3B Thinking 2507	Alibaba (Qwen)	$0.100	$0.100	262K	7
nvidia-nemotron-nano-9b-v2	NVIDIA	$0.040	$0.160	131K	6
Qwen/Qwen3.5-9B	Alibaba (Qwen)	$0.050	$0.150	262K	6
Qwen/Qwen3-VL-30B-A3B-Thinking	Alibaba (Qwen)	$0.100	$0.100	262K	6
Qwen/Qwen3-VL-30B-A3B-Instruct	Alibaba (Qwen)	$0.100	$0.100	262K	6
Qwen/Qwen3-VL-8B-Instruct	Alibaba (Qwen)	$0.100	$0.100	262K	5
deepseek-ai/DeepSeek-R1-Distill-Qwen-14B	DeepSeek	$0.100	$0.100	131K	4
Reka Edge	kilo	$0.100	$0.100	16K	1
Llama 3.1 8B	Meta	$0.100	$0.100	32K	1
GPT OSS 120B	synthetic	$0.100	$0.100	128K	1
Ministral 8B (latest)	Mistral	$0.100	$0.100	128K	1
GLM-4.6V-Flash	Z.AI / Zhipu	$0.020	$0.210	128K	3
Qwen Doc Turbo	Alibaba (Qwen)	$0.087	$0.144	131K	1
Mistral Small 3.2 24B Instruct	Mistral	$0.060	$0.180	96K	3
Qwen Turbo	Alibaba (Qwen)	$0.050	$0.200	1M	6
nemotron-3-nano-30b-a3b	NVIDIA	$0.050	$0.200	131K	6
GPT OSS 20B	fireworks-ai	$0.050	$0.200	131K	1
GPT OSS 20B	databricks	$0.050	$0.200	131K	1
GPT OSS Safeguard 20B	OpenAI	$0.070	$0.200	128K	6
Qwen/Qwen2.5-VL-32B-Instruct	Alibaba (Qwen)	$0.050	$0.220	131K	6
GPT OSS 20B	frogbot	$0.070	$0.200	131K	1
Llama-3.3-70B-Instruct	Meta	$0.050	$0.230	128K	22
Qwen 2.5 72B Instruct	Alibaba (Qwen)	$0.062	$0.231	32K	3
Mistral Nemo	Mistral	$0.150	$0.150	128K	8
Llama 3.3 Nemotron Super 49B v1.5	NVIDIA	$0.050	$0.250	131K	3
Llama 3.3 Nemotron Super 49B v1	NVIDIA	$0.150	$0.150	131K	2
Pixtral 12B	Mistral	$0.150	$0.150	128K	2

Top 60 von 660 angezeigt. Im vollständigen Verzeichnis weiter filtern.

Frequently asked questions

How many AI models support Tool Calling?

660 canonical models in our database currently support Tool Calling. The list is regenerated on every data refresh, so it always reflects the latest model releases from models.dev.

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 K2.5 (45), MiniMax-M2.5 (40), GLM-5 (38).

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 pulls models.dev once a day, regenerates the canonical model list, and rebuilds these pages so newly released models appear within 24 hours.

Top models with this capability

Voxtral Small 24B 2507$0.00 in / $0.00 out
Meta-Llama-3.1-8B-Instruct$0.02 in / $0.03 out
Llama 3.1 8B Turbo$0.02 in / $0.03 out
Mistral Nemo Instruct 2407$0.02 in / $0.04 out
Hermes 4 14B$0.01 in / $0.05 out

Other capabilities

Best-of lists you might also want

Pricing comparisons

Vendors in this list

Zuletzt aktualisiert: 2026-05-12

Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.

Data is sourced from models.dev and normalized for comparison. Prices and capabilities may change. Always verify critical production decisions with the provider's official documentation.