AIModelPort Hub

Ranking

Best Function Calling Models

Rank models for tool calling, agent workflows, and structured outputs.

Answer Summary / AI 摘要

Updated 2026-09-02

Best Function Calling Models 使用 reasoning 作为核心指标,对 190 个 AI 模型进行排序。English summary: this ranking compares AI models by reasoning; the current top result is GPT-5.1.

  • Primary metric: reasoning
  • Models included: 190
  • Top result: GPT-5.1

ModelPort API

测试榜单中的候选模型

看完 Best Function Calling Models 后,可以进入 ModelPort API 测试排名靠前的模型,并用自己的任务集比较质量、速度和成本。

Rank模型目录评分*
#1GPT-5.1

OpenAI

98/100
#2Claude 4.6 Sonnet

Anthropic

94/100
#3Gemini 3 Pro

Google

92/100
#4DeepSeek V4

DeepSeek

90/100
#5Qwen 3.5 Max

Alibaba Cloud

86/100
#6Mistral Large 3

Mistral AI

84/100
#7AionLabs: Aion-2.0

Other

76/100
#8AionLabs: Aion-3.0

Other

76/100
#9AionLabs: Aion-3.0-Mini

Other

76/100
#10Amazon: Nova 2 Lite

Other

76/100
#11Amazon: Nova Premier 1.0

Other

76/100
#12Anthropic Claude Haiku Latest

Anthropic

76/100
#13Anthropic Claude Sonnet Latest

Anthropic

76/100
#14Anthropic: Claude Fable 5

Anthropic

76/100
#15Anthropic: Claude Fable 5.1

Anthropic

76/100
#16Anthropic: Claude Fable Latest

Anthropic

76/100
#17Anthropic: Claude Haiku 4.5

Anthropic

76/100
#18Anthropic: Claude Opus 4.5

Anthropic

76/100
#19Anthropic: Claude Opus 4.6

Anthropic

76/100
#20Anthropic: Claude Opus 4.7

Anthropic

76/100
#21Anthropic: Claude Opus 4.8

Anthropic

76/100
#22Anthropic: Claude Opus 4.8 (Fast)

Anthropic

76/100
#23Anthropic: Claude Opus Latest

Anthropic

76/100
#24Anthropic: Claude Sonnet 4.6

Anthropic

76/100
#25Anthropic: Claude Sonnet 5

Anthropic

76/100
#26Arcee AI: Trinity Large Thinking

Other

76/100
#27Arcee AI: Trinity Mini

Other

76/100
#28Auto Router (Beta)

Other

76/100
#29ByteDance Seed: Seed 1.6

Other

76/100
#30ByteDance Seed: Seed 1.6 Flash

Other

76/100
#31ByteDance Seed: Seed 2.1 Turbo

Other

76/100
#32ByteDance Seed: Seed-2.0-Code

Other

76/100
#33ByteDance Seed: Seed-2.0-Lite

Other

76/100
#34ByteDance Seed: Seed-2.0-Mini

Other

76/100
#35Claude Opus 5

Anthropic

76/100
#36Claude Opus 5 (Fast)

Anthropic

76/100
#37DeepSeek V4 Flash Latest

DeepSeek

76/100
#38DeepSeek: DeepSeek V3.2

DeepSeek

76/100
#39DeepSeek: DeepSeek V4 Flash 0423

DeepSeek

76/100
#40DeepSeek: DeepSeek V4 Flash 0731

DeepSeek

76/100
#41DeepSeek: DeepSeek V4 Flash Vision Exp

DeepSeek

76/100
#42DeepSeek: DeepSeek V4 Pro 0423

DeepSeek

76/100
#43DeepSeek: DeepSeek V4 Pro 0813

DeepSeek

76/100
#44EssentialAI: Rnj 1 Instruct

Other

76/100
#45Google Gemini Flash Latest

Google

76/100
#46Google Gemini Pro Latest

Google

76/100
#47Google: Gemini 3 Flash Preview

Google

76/100
#48Google: Gemini 3.1 Flash Lite

Google

76/100
#49Google: Gemini 3.1 Flash Lite Preview

Google

76/100
#50Google: Gemini 3.1 Pro Preview

Google

76/100
#51Google: Gemini 3.1 Pro Preview Custom Tools

Google

76/100
#52Google: Gemini 3.5 Flash

Google

76/100
#53Google: Gemini 3.5 Flash Lite

Google

76/100
#54Google: Gemini 3.6 Flash

Google

76/100
#55Google: Gemini 3.7 Flash

Google

76/100
#56Google: Gemma 4 26B A4B

Google

76/100
#57Google: Gemma 4 31B

Google

76/100
#58IBM: Granite 4.1 8B

Other

76/100
#59IBM: Granite 4.2 8B

Other

76/100
#60Inception: Mercury 2

Other

76/100
#61Inception: Mercury 2.5 Preview

Other

76/100
#62Kwaipilot: KAT-Coder-Air V2.5

Other

76/100
#63Kwaipilot: KAT-Coder-Pro V2

Other

76/100
#64Kwaipilot: KAT-Coder-Pro V2.5

Other

76/100
#65Ling-3.0-flash

Other

76/100
#66Meituan: LongCat 2.0

Other

76/100
#67Meta: Muse Glimmer 30B

Other

76/100
#68Meta: Muse Spark 1.1

Other

76/100
#69Meta: Muse Spark 1.2

Other

76/100
#70Meta: Muse Spark 1.2 Contributor

Other

76/100
#71MiniMax: MiniMax M2

Other

76/100
#72MiniMax: MiniMax M2.1

Other

76/100
#73MiniMax: MiniMax M2.5

Other

76/100
#74MiniMax: MiniMax M2.7

Other

76/100
#75MiniMax: MiniMax M3

Other

76/100
#76Mistral: Devstral 2 2512

Mistral AI

76/100
#77Mistral: Ministral 3 14B 2512

Mistral AI

76/100
#78Mistral: Ministral 3 3B 2512

Mistral AI

76/100
#79Mistral: Ministral 3 8B 2512

Mistral AI

76/100
#80Mistral: Mistral Large 3 2512

Mistral AI

76/100
#81Mistral: Mistral Medium 3.5

Mistral AI

76/100
#82Mistral: Mistral Small 4

Mistral AI

76/100
#83Mistral: Voxtral Small 24B 2507

Mistral AI

76/100
#84MoonshotAI Kimi Latest

Other

76/100
#85MoonshotAI: Kimi K2 Thinking

Other

76/100
#86MoonshotAI: Kimi K2.5

Other

76/100
#87MoonshotAI: Kimi K2.6

Other

76/100
#88MoonshotAI: Kimi K2.7 Code

Other

76/100
#89MoonshotAI: Kimi K3

Other

76/100
#90NVIDIA: Llama 3.3 Nemotron Super 49B V1.5

Other

76/100
#91NVIDIA: Nemotron 3 Nano 30B A3B

Other

76/100
#92NVIDIA: Nemotron 3 Super

Other

76/100
#93NVIDIA: Nemotron 3 Ultra

Other

76/100
#94NVIDIA: Nemotron 3.5 Lightning

Other

76/100
#95Nex AGI: DeepSeek V3.1 Nex N1

Other

76/100
#96Nex AGI: Nex-N2-Mini

Other

76/100
#97Nex AGI: Nex-N2-Pro

Other

76/100
#98OpenAI GPT Latest

OpenAI

76/100
#99OpenAI GPT Mini Latest

OpenAI

76/100
#100OpenAI: GPT Audio

OpenAI

76/100
#101OpenAI: GPT Audio Mini

OpenAI

76/100
#102OpenAI: GPT Chat Latest

OpenAI

76/100
#103OpenAI: GPT-5.1 Chat

OpenAI

76/100
#104OpenAI: GPT-5.1-Codex

OpenAI

76/100
#105OpenAI: GPT-5.1-Codex-Max

OpenAI

76/100
#106OpenAI: GPT-5.1-Codex-Mini

OpenAI

76/100
#107OpenAI: GPT-5.2

OpenAI

76/100
#108OpenAI: GPT-5.2 Chat

OpenAI

76/100
#109OpenAI: GPT-5.2-Codex

OpenAI

76/100
#110OpenAI: GPT-5.3 Chat

OpenAI

76/100
#111OpenAI: GPT-5.3-Codex

OpenAI

76/100
#112OpenAI: GPT-5.4

OpenAI

76/100
#113OpenAI: GPT-5.4 Mini

OpenAI

76/100
#114OpenAI: GPT-5.4 Nano

OpenAI

76/100
#115OpenAI: GPT-5.5

OpenAI

76/100
#116OpenAI: GPT-5.6 Luna

OpenAI

76/100
#117OpenAI: GPT-5.6 Luna Pro

OpenAI

76/100
#118OpenAI: GPT-5.6 Sol

OpenAI

76/100
#119OpenAI: GPT-5.6 Sol Pro

OpenAI

76/100
#120OpenAI: GPT-5.6 Terra

OpenAI

76/100
#121OpenAI: GPT-5.6 Terra Pro

OpenAI

76/100
#122OpenAI: gpt-oss-safeguard-20b

OpenAI

76/100
#123OpenAI: o3 Deep Research

OpenAI

76/100
#124OpenAI: o4 Mini Deep Research

OpenAI

76/100
#125Poolside: Laguna M.1

Other

76/100
#126Poolside: Laguna S 2.1

Other

76/100
#127Poolside: Laguna XS 2.1

Other

76/100
#128Poolside: Laguna XS.2

Other

76/100
#129Prime Intellect: INTELLECT-3

Other

76/100
#130Qwen: Qwen3 Coder Next

Alibaba Cloud

76/100
#131Qwen: Qwen3 Max Thinking

Alibaba Cloud

76/100
#132Qwen: Qwen3 VL 30B A3B Thinking

Alibaba Cloud

76/100
#133Qwen: Qwen3 VL 32B Instruct

Alibaba Cloud

76/100
#134Qwen: Qwen3 VL 8B Instruct

Alibaba Cloud

76/100
#135Qwen: Qwen3 VL 8B Thinking

Alibaba Cloud

76/100
#136Qwen: Qwen3.5 397B A17B

Alibaba Cloud

76/100
#137Qwen: Qwen3.5 Plus 2026-02-15

Alibaba Cloud

76/100
#138Qwen: Qwen3.5 Plus 2026-04-20

Alibaba Cloud

76/100
#139Qwen: Qwen3.5-122B-A10B

Alibaba Cloud

76/100
#140Qwen: Qwen3.5-27B

Alibaba Cloud

76/100
#141Qwen: Qwen3.5-35B-A3B

Alibaba Cloud

76/100
#142Qwen: Qwen3.5-9B

Alibaba Cloud

76/100
#143Qwen: Qwen3.5-Flash

Alibaba Cloud

76/100
#144Qwen: Qwen3.6 27B

Alibaba Cloud

76/100
#145Qwen: Qwen3.6 35B A3B

Alibaba Cloud

76/100
#146Qwen: Qwen3.6 Flash

Alibaba Cloud

76/100
#147Qwen: Qwen3.6 Max Preview

Alibaba Cloud

76/100
#148Qwen: Qwen3.6 Plus

Alibaba Cloud

76/100
#149Qwen: Qwen3.7 Flash

Alibaba Cloud

76/100
#150Qwen: Qwen3.7 Max

Alibaba Cloud

76/100
#151Qwen: Qwen3.7 Plus

Alibaba Cloud

76/100
#152Qwen: Qwen3.8 2.4T A95B

Alibaba Cloud

76/100
#153Qwen: Qwen3.8 27B

Alibaba Cloud

76/100
#154Qwen: Qwen3.8 Flash

Alibaba Cloud

76/100
#155Qwen: Qwen3.8 Max

Alibaba Cloud

76/100
#156Reka Edge

Other

76/100
#157Relace: Relace Search

Other

76/100
#158Sakana: Fugu Ultra

Other

76/100
#159Sakana: Sakana Namazu

Other

76/100
#160SpaceXAI: Grok 4.20

Other

76/100
#161SpaceXAI: Grok 4.3

Other

76/100
#162SpaceXAI: Grok 4.5

Other

76/100
#163SpaceXAI: Grok 4.6

Other

76/100
#164SpaceXAI: Grok Build 0.1

Other

76/100
#165StepFun: Step 3.5 Flash

Other

76/100
#166StepFun: Step 3.7 Flash

Other

76/100
#167Tencent: Hy3

Other

76/100
#168Tencent: Hy3 preview

Other

76/100
#169Tencent: Hy4 preview

Other

76/100
#170Upstage: Solar Pro 3

Other

76/100
#171Upstage: Solar Pro 4

Other

76/100
#172Xiaomi: MiMo-V2-Flash

Other

76/100
#173Xiaomi: MiMo-V2.5

Other

76/100
#174Xiaomi: MiMo-V2.5-Pro

Other

76/100
#175Z.ai: GLM 4.6V

Other

76/100
#176Z.ai: GLM 4.7

Other

76/100
#177Z.ai: GLM 4.7 Flash

Other

76/100
#178Z.ai: GLM 5

Other

76/100
#179Z.ai: GLM 5 Turbo

Other

76/100
#180Z.ai: GLM 5.1

Other

76/100
#181Z.ai: GLM 5.2

Other

76/100
#182Z.ai: GLM 5.3

Other

76/100
#183Z.ai: GLM 5.3 Flash

Other

76/100
#184Z.ai: GLM 5V Turbo

Other

76/100
#185Z.ai: GLM Flash Latest

Other

76/100
#186Z.ai: GLM Latest

Other

76/100
#187inclusionAI: Ling-2.6-1T

Other

76/100
#188inclusionAI: Ling-2.6-flash

Other

76/100
#189inclusionAI: Ring-2.6-1T

Other

76/100
#190xAI: Grok Latest

Other

76/100

* 目录评分用于候选模型排序,基于编辑判断与目录能力信号,不代表独立 benchmark 成绩。生产选型请使用自己的任务集复测。

Facts Table / Source Facts

字段含义、来源优先级与更新时间规则见数据与排名方法

Machine-readable facts for AI search engines and answer engines.
FieldValueNote
RankingBest Function Calling ModelsCatalog data
Primary metricreasoningCatalog data
Models included190Catalog data
MethodologySort by editorial catalog signals, then review price, context window, vision support, and function calling. The score is not an independent benchmark result.Catalog data
Updated2026-09-02Catalog data

排名方法

Sort by editorial catalog signals, then review price, context window, vision support, and function calling. The score is not an independent benchmark result.

FAQ

Best Function Calling Models 排名如何计算?+

排名由结构化目录数据和编辑判断生成,结合价格、能力信号、上下文长度和关键 API 能力;目录评分不是独立 benchmark 成绩。

这个排行榜适合直接做采购决策吗?+

它适合做候选列表,生产采购前仍需要用自己的任务样本做评估。

引用格式 / Citation Format

AI 搜索或研究型回答可以引用下面的稳定格式。

Plain text

ModelPort Hub. "Best Function Calling Models: AI Model Ranking." Updated 2026-09-02. https://modelporthub.com/rankings/best-function-calling-models

Markdown

[Best Function Calling Models: AI Model Ranking](https://modelporthub.com/rankings/best-function-calling-models) — ModelPort Hub, updated 2026-09-02.