Leaderboard

    Best AI for Agentic Tasks 2026.

    Find the best AI for agentic tasks and tool use. Ranked by OSWorld, ToolAthlon, MCP Atlas, tau-bench, BrowseComp, and more agent benchmarks.

    LLMImageVideoSpeechTranscription
    Claude Opus 4.6
    Anthropic
    		0.8%—0.7%—0.6%—
    Claude Sonnet 4.6
    Anthropic
    		0.7%—0.7%—0.6%—
    A
    Qwen3 VL 235B A22B InstructOSS
    Alibaba Cloud / Qwen Team
    		—0.6%0.7%———
    Claude Opus 4.5
    Anthropic
    		——0.7%—0.6%—
    Claude Sonnet 4.5
    Anthropic
    		——0.6%——0.9%
    Claude Haiku 4.5
    Anthropic
    		——0.5%———
    A
    Qwen3 VL 235B A22B ThinkingOSS
    Alibaba Cloud / Qwen Team
    		—0.6%0.4%———
    A
    Qwen3 VL 8B ThinkingOSS
    Alibaba Cloud / Qwen Team
    		—0.5%0.3%———
    A
    Qwen3 VL 8B InstructOSS
    Alibaba Cloud / Qwen Team
    		—0.5%0.3%———
    A
    Qwen3 VL 4B ThinkingOSS
    Alibaba Cloud / Qwen Team
    		—0.5%0.3%———
    A
    Qwen3 VL 30B A3B ThinkingOSS
    Alibaba Cloud / Qwen Team
    		—0.6%0.3%———
    A
    Qwen3 VL 30B A3B InstructOSS
    Alibaba Cloud / Qwen Team
    		—0.6%0.3%———
    A
    Qwen3 VL 4B InstructOSS
    Alibaba Cloud / Qwen Team
    		—0.6%0.3%———
    ChatGPT-4o Latest
    OpenAI
    		——————
    Claude 3 Haiku
    Anthropic
    		——————
    Claude 3 Opus
    Anthropic
    		——————
    Claude 3 Sonnet
    Anthropic
    		——————
    Claude 3.5 Haiku
    Anthropic
    		—————0.5%
    Claude 3.5 Sonnet
    Anthropic
    		—————0.7%
    Claude 3.7 Sonnet
    Anthropic
    		—————0.8%
    Claude Opus 4
    Anthropic
    		—————0.8%
    Claude Opus 4.1
    Anthropic
    		—————0.8%
    Claude Sonnet 4
    Anthropic
    		—————0.8%
    C
    Command R+OSS
    Cohere
    		——————
    DeepSeek R1 Distill Llama 70BOSS
    DeepSeek
    		——————
    DeepSeek R1 Distill Qwen 32BOSS
    DeepSeek
    		——————
    DeepSeek-R1OSS
    DeepSeek
    		——————
    DeepSeek-R1-0528OSS
    DeepSeek
    		0.1%—————
    DeepSeek-V2.5OSS
    DeepSeek
    		——————
    DeepSeek-V3OSS
    DeepSeek
    		——————
    DeepSeek-V3 0324OSS
    DeepSeek
    		——————
    DeepSeek-V3.1OSS
    DeepSeek
    		0.3%—————
    DeepSeek-V3.2 (Non-thinking)OSS
    DeepSeek
    		——————
    DeepSeek-V3.2-ExpOSS
    DeepSeek
    		0.4%—————
    Devstral Medium
    Mistral AI
    		——————
    Devstral Small 1.1OSS
    Mistral AI
    		——————
    B
    ERNIE 4.5
    Baidu
    		——————
    Gemini 1.0 Pro
    Google
    		——————
    Gemini 1.5 Flash
    Google
    		——————
    Gemini 1.5 Flash 8B
    Google
    		——————
    Gemini 1.5 Pro
    Google
    		——————
    Gemini 2.0 Flash
    Google
    		——————
    Gemini 2.0 Flash-Lite
    Google
    		——————
    Gemini 2.5 Flash
    Google
    		——————
    Gemini 2.5 Flash-LiteOSS
    Google
    		——————
    Gemini 2.5 Pro
    Google
    		——————
    Gemini 2.5 Pro Preview 06-05
    Google
    		——————
    Gemini 3 Flash
    Google
    		—0.7%—0.5%0.6%—
    Gemini 3 Pro
    Google
    		—0.7%————
    Gemini 3.1 Flash-Lite
    Google
    		——————
    Showing 1–50 of 174 models

    Building with these APIs?

    Get 10+ Next.js AI templates with auth, payments, and more.

    Get Templates — $249

    All Large Language Models

    OpenAI

    39 models
    GPT-5.4·GPT-5.3 Chat·GPT-5.3 Codex·GPT-5.2 Codex·GPT-5.2·GPT-5.2 Pro·GPT-5.1 Codex·GPT-5.1·GPT-5.1 Codex High·GPT-5.1 Codex Mini·GPT-5.1 High·GPT-5.1 Instant·GPT-5.1 Medium·GPT-5.1 Thinking·GPT-5·GPT-5 High·GPT-5 Medium·GPT-5 mini·GPT-5 nano·GPT OSS 120B·GPT OSS 120B High·GPT OSS 20B·GPT OSS 20B High·o3-pro·o3·o4-mini·GPT-4.1·GPT-4.1 mini·GPT-4.1 nano·o3-mini·o1·o1-mini·o1-preview·GPT-4o·GPT-4o mini·ChatGPT-4o Latest·GPT-4 Turbo·GPT-4·GPT-3.5 Turbo

    Anthropic

    14 models
    Claude Sonnet 4.6·Claude Opus 4.6·Claude Opus 4.5·Claude Haiku 4.5·Claude Sonnet 4.5·Claude Opus 4.1·Claude Opus 4·Claude Sonnet 4·Claude 3.7 Sonnet·Claude 3.5 Haiku·Claude 3.5 Sonnet·Claude 3 Haiku·Claude 3 Opus·Claude 3 Sonnet

    Google

    18 models
    Gemini 3.1 Flash-Lite·Gemini 3.1 Pro·Gemini 3 Flash·Gemini 3 Pro·Gemma 3n E4B Instructed·Gemini 2.5 Flash-Lite·Gemini 2.5 Pro Preview 06-05·Gemini 2.5 Flash·Gemini 2.5 Pro·Gemma 3 12B·Gemma 3 27B·Gemma 3 4B·Gemini 2.0 Flash-Lite·Gemini 2.0 Flash·Gemini 1.5 Flash·Gemini 1.5 Pro·Gemini 1.5 Flash 8B·Gemini 1.0 Pro

    xAI

    12 models
    Grok-4.1·Grok-4.1 Fast Non-Reasoning·Grok-4.1 Fast Reasoning·Grok-4.1 Thinking·Grok 4 Fast·Grok Code Fast 1·Grok-4 Fast Non-Reasoning·Grok-4 Fast Reasoning·Grok-4·Grok-3·Grok-3 Mini·Grok-2

    DeepSeek

    10 models
    DeepSeek-V3.2 (Non-thinking)·DeepSeek-V3.2-Exp·DeepSeek-R1-0528·DeepSeek-V3 0324·DeepSeek R1 Distill Llama 70B·DeepSeek R1 Distill Qwen 32B·DeepSeek-R1·DeepSeek-V3.1·DeepSeek-V3·DeepSeek-V2.5

    MiniMax

    4 models
    MiniMax M2.5·MiniMax M2.1·MiniMax M2·MiniMax M1 80K

    AI21 Labs

    2 models
    Jamba 1.5 Large·Jamba 1.5 Mini

    Alibaba Cloud / Qwen Team

    22 models
    Qwen3.5-397B-A17B·Qwen3 Max·Qwen3 VL 235B A22B Instruct·Qwen3 VL 235B A22B Thinking·Qwen3 VL 30B A3B Instruct·Qwen3 VL 30B A3B Thinking·Qwen3 VL 4B Instruct·Qwen3 VL 4B Thinking·Qwen3 VL 8B Instruct·Qwen3 VL 8B Thinking·Qwen3-Next-80B-A3B-Instruct·Qwen3-Next-80B-A3B-Thinking·Qwen3-235B-A22B-Thinking-2507·Qwen3-235B-A22B-Instruct-2507·Qwen3 235B A22B·Qwen3 30B A3B·Qwen3 32B·Qwen3-Coder·QwQ-32B-Preview·Qwen2.5 72B Instruct·Qwen2.5 7B Instruct·Qwen2.5-Coder 32B Instruct

    Amazon

    3 models
    Nova Lite·Nova Micro·Nova Pro

    Baidu

    1 models
    ERNIE 4.5

    Cohere

    1 models
    Command R+

    IBM

    1 models
    Granite 3.3 8B Instruct

    Inception

    1 models
    Mercury 2

    LG AI Research

    1 models
    K-EXAONE-236B-A23B

    Meituan

    4 models
    LongCat-Flash-Lite·LongCat-Flash-Thinking-2601·LongCat-Flash-Thinking·LongCat-Flash-Chat

    Meta

    9 models
    Llama 4 Maverick·Llama 4 Scout·Llama 3.3 70B Instruct·Llama 3.2 11B Instruct·Llama 3.2 3B Instruct·Llama 3.2 90B Instruct·Llama 3.1 405B Instruct·Llama 3.1 70B Instruct·Llama 3.1 8B Instruct

    Microsoft

    3 models
    Phi-4-multimodal-instruct·Phi 4·Phi-3.5-mini-instruct

    Mistral AI

    15 models
    Min istral 3 (3B Reasoning 2512)·Ministral 3 (14B Reasoning 2512)·Ministral 3 (8B Reasoning 2512)·Mistral Large 3 (675B Instruct 2512)·Mistral Large 3·Devstral Small 1.1·Devstral Medium·Mistral Small 3.1 24B Base·Mistral Small 3 24B Instruct·Pixtral Large·Ministral 8B Instruct·Mistral Small·Pixtral-12B·Mistral Large 2·Mistral NeMo Instruct

    Moonshot AI

    4 models
    Kimi K2.5·Kimi K2 0905·Kimi K2-Thinking-0905·Kimi K2 Instruct

    Nous Research

    1 models
    Hermes 3 70B

    NVIDIA

    1 models
    Nemotron 3 Nano (30B A3B)

    StepFun

    1 models
    Step-3.5-Flash

    Xiaomi

    1 models
    MiMo-V2-Flash

    Zhipu AI

    6 models
    GLM-5·GLM-4.7-Flash·GLM-4.7·GLM-4.6·GLM-4.5V·GLM-4.5
    Best AI for Agentic Tasks 2026 — AI Agent Leaderboard | AnotherWrapper