LLM Agent Benchmark Leaderboard

Name: LLM Agent Benchmark Leaderboard
Creator: DataLearner
License: https://creativecommons.org/licenses/by/4.0/

This page provides the LLM Agent benchmark leaderboard, covering Aider-Polyglot, τ²-Bench, Terminal Bench 2.0, Tool Decathlon, and OSWorld-Verified. Compare GPT, Claude, Qwen, and DeepSeek on tool use, task planning, and autonomous execution.

Updated on 2026-07-28 08:43:41

As of 2026-07, this page covers Aider-Polyglot, τ²-Bench, Terminal Bench 2.0, Tool Decathlon and related benchmarks for LLM Agent Benchmark Leaderboard, making it straightforward to compare within the same task family.

Click any model name to check context length, licensing, and pricing on its detail page. See Data Methodology for scoring details.

Benchmark

Agent能力评测Aider-Polyglot τ²-Bench

AI Agent - 工具使用Terminal Bench 2.0 Tool Decathlon OSWorld-Verified

More Benchmarks

Model Size:All 3B and below 7B 13B 34B 65B 100B and above

Model Type:All Reasoning Models Foundation Models Instruction/Chat Models Coding Models

Source:All Open Source Closed Source

Origin:All China

Model release cutoff:

Top picks

Ranked by τ²-Bench

Current SOTA

Claude Opus 4.6

Anthropic

91.89τ²-Bench

View model

Best Open-Source

GLM-5

智谱AI

89.70τ²-Bench−2.19

View model

Best China-Made

GLM-5

智谱AI

89.70τ²-Bench−2.19

View model

LLM Performance Results

Data source: DataLearnerAI

Click any row to open the model page. Tick the checkboxes to compare up to 4 models side by side.

Rank	Model						License
	Claude Opus 4.6 Extended ThinkingTools Anthropic	—	91.89	65.40	—	72.70	Proprietary	Details
	Gemini 3.1 Pro Preview Thinking EnabledTools Google Deep Mind	—	90.80	68.50	—	—	Proprietary	Details
	Gemini 3.0 Flash Thinking EnabledTools Google Deep Mind	—	90.20	47.60	—	—	Proprietary	Details
4	GLM-5 Thinking EnabledTools 智谱AI	—	89.70	61.10	—	—	Free commercial	Details
5	Step 3.5 Flash Thinking EnabledTools StepFunAI	—	88.20	51.00	—	—	Free commercial	Details
6	GLM-4.7 Thinking EnabledTools 智谱AI	—	87.40	41.00	—	—	Free commercial	Details
7	Qwen3.5-397B-A17B Thinking EnabledTools 阿里巴巴	—	86.70	52.50	38.30	62.20	Free commercial	Details
8	Gemini 3.0 Pro (Preview 11-2025) Thinking EnabledTools Google Deep Mind	—	85.40	54.20	—	—	Proprietary	Details
9	Claude Sonnet 4.5 Thinking EnabledTools Anthropic	—	84.70	42.80	—	61.40	Proprietary	Details
10	Grok 4.1 Fast Thinking EnabledTools xAI	—	82.71	—	—	—	Proprietary	Details
11	Qwen3-Max-Thinking Thinking EnabledTools 阿里巴巴	—	82.10	—	—	—	Proprietary	Details
12	GPT-5.2 Thinking Level · Extra HighTools OpenAI	—	82.00	—	—	—	Proprietary	Details
13	Opus 4.5 Extended ThinkingTools Anthropic	—	81.99	59.30	—	—	Proprietary	Details
14	DeepSeek V3.2 Thinking EnabledTools DeepSeek-AI	—	80.30	46.40	—	—	Free commercial	Details
15	GPT-5 Thinking EnabledTools OpenAI	—	80.00	—	—	—	Proprietary	Details
16	GLM-4.7-Flash Thinking EnabledTools 智谱AI	—	79.50	—	—	—	Free commercial	Details
17	Qwen3.5-27B Thinking EnabledTools 阿里巴巴	—	79.00	41.60	—	56.20	Free commercial	Details
18	ERNIE 5.0 Thinking Enabled 百度	—	78.79	—	—	—	Proprietary	Details
19	MiniMax M2 Thinking EnabledTools MiniMaxAI	—	77.20	—	—	—	Free commercial	Details
20	Gemma 4 31B Thinking EnabledTools DeepMind	—	76.90	—	—	—	Free commercial	Details
21	GLM-4.6 Thinking EnabledTools 智谱AI	—	75.90	—	—	—	Free commercial	Details
22	Qwen3 Max (Preview) Standard ModeTools 阿里巴巴	—	74.00	—	—	—	Proprietary	Details
23	Claude Opus 4 Thinking EnabledTools Anthropic	—	72.50	—	—	—	Proprietary	Details
24	Qwen3 Max (Preview) Thinking EnabledTools 阿里巴巴	—	72.00	—	—	—	Proprietary	Details
25	Claude Sonnet 4.5 Standard ModeTools Anthropic	—	71.00	—	—	—	Proprietary	Details
26	Gemma 4 26B A4B Thinking EnabledTools DeepMind	—	68.20	—	—	—	Free commercial	Details
27	DeepSeek V3.2-Exp Thinking EnabledTools DeepSeek-AI	—	66.70	—	—	—	Free commercial	Details
28	Kimi K2 Thinking EnabledTools Moonshot AI	—	64.30	—	—	—	Free commercial	Details
29	Kimi K2 Standard ModeTools Moonshot AI	—	64.30	—	—	—	Free commercial	Details
30	Claude Sonnet 3.7 Thinking EnabledTools Anthropic	—	61.80	—	—	28.00	Proprietary	Details
31	OpenAI o4 - mini Thinking EnabledTools OpenAI	—	56.90	—	—	—	Proprietary	Details
32	GPT-4.1 Standard ModeTools OpenAI	—	54.70	—	—	—	Proprietary	Details
33	GPT-4.1 mini Standard ModeTools OpenAI	—	53.00	—	—	—	Proprietary	Details
34	Claude Sonnet 4 Standard ModeTools Anthropic	—	52.00	—	—	—	Proprietary	Details
35	Qwen3-30B-A3B-2507 Thinking EnabledTools 阿里巴巴	—	49.00	—	—	—	Free commercial	Details
36	GPT OSS 20B Thinking EnabledTools OpenAI	—	47.70	—	—	—	Free commercial	Details
37	Gemma 4 E4B Thinking Enabled DeepMind	—	42.20	—	—	—	Free commercial	Details
38	DeepSeek-V3-0324 Standard ModeTools DeepSeek-AI	—	38.80	—	—	—	Free commercial	Details
39	DeepSeek-V3.1 Terminus Thinking EnabledTools DeepSeek-AI	—	37.00	—	—	—	Free commercial	Details
40	DeepSeek-V3.1 Terminus Standard ModeTools DeepSeek-AI	—	37.00	—	—	—	Free commercial	Details
41	Qwen3-235B-A22B Thinking EnabledTools 阿里巴巴	—	34.40	—	—	—	Free commercial	Details
42	Haiku 4.5 Standard ModeTools Anthropic	—	33.00	—	—	—	Proprietary	Details
43	Gemma 4 E2B Thinking Enabled DeepMind	—	24.50	—	—	—	Free commercial	Details
44	GPT-5 Thinking Enabled OpenAI	88.00	—	—	—	—	Proprietary	Details
45	GPT-5 Thinking Enabled OpenAI	86.70	—	—	—	—	Proprietary	Details
46	o3-pro Thinking Enabled OpenAI	84.90	—	—	—	—	Proprietary	Details
47	Gemini 2.5-Pro Thinking Enabled Google Deep Mind	83.10	—	—	—	—	Proprietary	Details
48	OpenAI o3 Thinking Enabled OpenAI	81.30	—	—	—	—	Proprietary	Details
49	GPT-5 Thinking Enabled OpenAI	81.30	—	—	—	—	Proprietary	Details
50	GPT-4.1 nano OpenAI	8.90	—	—	—	—	Proprietary	Details

Claude Opus 4.6 Anthropic

Extended ThinkingTools

Aider-Polyglot—

τ²-Bench91.89

Terminal Bench 2.065.40

Tool Decathlon—

OSWorld-Verified72.70

Proprietary

Gemini 3.1 Pro Preview Google Deep Mind

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench90.80

Terminal Bench 2.068.50

Tool Decathlon—

OSWorld-Verified—

Proprietary

Gemini 3.0 Flash Google Deep Mind

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench90.20

Terminal Bench 2.047.60

Tool Decathlon—

OSWorld-Verified—

Proprietary

GLM-5 智谱AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench89.70

Terminal Bench 2.061.10

Tool Decathlon—

OSWorld-Verified—

Free commercial

Step 3.5 Flash StepFunAI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench88.20

Terminal Bench 2.051.00

Tool Decathlon—

OSWorld-Verified—

Free commercial

GLM-4.7 智谱AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench87.40

Terminal Bench 2.041.00

Tool Decathlon—

OSWorld-Verified—

Free commercial

Qwen3.5-397B-A17B 阿里巴巴

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench86.70

Terminal Bench 2.052.50

Tool Decathlon38.30

OSWorld-Verified62.20

Free commercial

Gemini 3.0 Pro (Preview 11-2025)Google Deep Mind

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench85.40

Terminal Bench 2.054.20

Tool Decathlon—

OSWorld-Verified—

Proprietary

Claude Sonnet 4.5 Anthropic

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench84.70

Terminal Bench 2.042.80

Tool Decathlon—

OSWorld-Verified61.40

Proprietary

Grok 4.1 Fast xAI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench82.71

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Qwen3-Max-Thinking 阿里巴巴

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench82.10

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GPT-5.2 OpenAI

Thinking Level · Extra HighTools

Aider-Polyglot—

τ²-Bench82.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Opus 4.5 Anthropic

Extended ThinkingTools

Aider-Polyglot—

τ²-Bench81.99

Terminal Bench 2.059.30

Tool Decathlon—

OSWorld-Verified—

Proprietary

DeepSeek V3.2 DeepSeek-AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench80.30

Terminal Bench 2.046.40

Tool Decathlon—

OSWorld-Verified—

Free commercial

GPT-5 OpenAI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench80.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GLM-4.7-Flash 智谱AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench79.50

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Qwen3.5-27B 阿里巴巴

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench79.00

Terminal Bench 2.041.60

Tool Decathlon—

OSWorld-Verified56.20

Free commercial

ERNIE 5.0 百度

Thinking Enabled

Aider-Polyglot—

τ²-Bench78.79

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

MiniMax M2 MiniMaxAI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench77.20

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Gemma 4 31B DeepMind

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench76.90

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

GLM-4.6 智谱AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench75.90

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Qwen3 Max (Preview)阿里巴巴

Standard ModeTools

Aider-Polyglot—

τ²-Bench74.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Claude Opus 4 Anthropic

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench72.50

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Qwen3 Max (Preview)阿里巴巴

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench72.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Claude Sonnet 4.5 Anthropic

Standard ModeTools

Aider-Polyglot—

τ²-Bench71.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Gemma 4 26B A4B DeepMind

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench68.20

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

DeepSeek V3.2-Exp DeepSeek-AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench66.70

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Kimi K2 Moonshot AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench64.30

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Kimi K2 Moonshot AI

Standard ModeTools

Aider-Polyglot—

τ²-Bench64.30

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Claude Sonnet 3.7 Anthropic

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench61.80

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified28.00

Proprietary

OpenAI o4 - mini OpenAI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench56.90

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GPT-4.1 OpenAI

Standard ModeTools

Aider-Polyglot—

τ²-Bench54.70

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GPT-4.1 mini OpenAI

Standard ModeTools

Aider-Polyglot—

τ²-Bench53.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Claude Sonnet 4 Anthropic

Standard ModeTools

Aider-Polyglot—

τ²-Bench52.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Qwen3-30B-A3B-2507 阿里巴巴

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench49.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

GPT OSS 20B OpenAI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench47.70

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Gemma 4 E4B DeepMind

Thinking Enabled

Aider-Polyglot—

τ²-Bench42.20

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

DeepSeek-V3-0324 DeepSeek-AI

Standard ModeTools

Aider-Polyglot—

τ²-Bench38.80

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

DeepSeek-V3.1 Terminus DeepSeek-AI

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench37.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

DeepSeek-V3.1 Terminus DeepSeek-AI

Standard ModeTools

Aider-Polyglot—

τ²-Bench37.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Qwen3-235B-A22B 阿里巴巴

Thinking EnabledTools

Aider-Polyglot—

τ²-Bench34.40

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

Haiku 4.5 Anthropic

Standard ModeTools

Aider-Polyglot—

τ²-Bench33.00

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Gemma 4 E2B DeepMind

Thinking Enabled

Aider-Polyglot—

τ²-Bench24.50

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Free commercial

GPT-5 OpenAI

Thinking Enabled

Aider-Polyglot88.00

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GPT-5 OpenAI

Thinking Enabled

Aider-Polyglot86.70

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

o3-pro OpenAI

Thinking Enabled

Aider-Polyglot84.90

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Gemini 2.5-Pro Google Deep Mind

Thinking Enabled

Aider-Polyglot83.10

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

OpenAI o3 OpenAI

Thinking Enabled

Aider-Polyglot81.30

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GPT-5 OpenAI

Thinking Enabled

Aider-Polyglot81.30

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

GPT-4.1 nano OpenAI

Aider-Polyglot8.90

τ²-Bench—

Terminal Bench 2.0—

Tool Decathlon—

OSWorld-Verified—

Proprietary

Sort by:

Showing 50 of 149 modelsView τ²-Bench benchmark page