Model Menu
Models
Compare
Leaderboard
Benchmarks
⌘K
Top
Chat
Coding
Agents
Writing
Vision
Image
Video
Speech
Embeddings
Decision
Music
Filters
Best for Agents & Tool Use
All models
2292
Chat & Knowledge
558
Coding
381
Agents & Tool Use
541
Writing & Language
360
Vision & Documents
395
Image Generation & Editing
183
Video Generation
148
Speech & Voice
358
Embeddings & Retrieval
671
Decision models
118
Music & Sound
29
AutomationBench-AA
GDPval-AA 2.1
Toolathlon-Verified
MCP-Atlas Public
MCP-Atlas full 1000
τ²-bench
BFCL v4 (past run, native function calls)
All benchmarks
83
Agents & Tool Use benchmarks
83
×
All
83
Agents
41
Health & law
1
Documents
1
Benchmark
By
Leader
Models
Agents
28
AutomationBench-AA
AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
AA
Gemini 4 Argon
77.5%
133
AutomationBench
Zapier run
AutomationBench · Zapier run: End-to-end completion of business workflows using real SaaS applications
Zapier
Gemini 4 Argon
51.3%
2
+1 versions & slices
AutomationBench
Reported by the lab with its own setup
Lab
GLM 5.3
62.2
7
GDPval-AA (work tasks)
2.1
GDPval-AA (work tasks) 2.1: Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work
AA
Claude Opus 5.5
68.3
175
Toolathlon-Verified
Toolathlon-Verified: Toolathlon-Verified as reported in this launch evaluation; the post is the authority for the recorded task and grading scope
GLM 5.3 Flash
78.4%
6
MCP-Atlas Public
MCPAtlas Public: MCPAtlas Public as reported in this launch evaluation; the post is the authority for the recorded task and grading scope
Muse Spark
82.4%
16
MCP-Atlas
full 1000
MCP-Atlas · full 1000: Tool orchestration across MCP servers on the stated task split
Scale
Muse Spark
82.2%
16
τ³-Banking
AA
τ³-Banking · AA: Agents retrieve banking knowledge and execute tools to resolve customer-support workflows
AA
Grok 4.6
50.7%
134
τ³-bench
banking · 1.0.1
τ-bench official · 3-banking-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Sierra
Claude Opus 5
48.7%
21
AA Analyst Agent
AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings
AA
Claude Fable 5.1
57.5%
43
ITBench SRE
ITBench SRE: IT incident tasks: diagnose and fix broken systems like a site-reliability engineer
AA
GPT-5.6 Sol
56.2%
37
APEX-Agents
APEX-Agents: Agents complete long-horizon professional-services tasks across realistic applications
AA
Qwen3.8 Max (0902)
42.4%
29
Terminal-Bench Hard
Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal
AA
GPT-5.6 Sol
65.9%
239
Terminal-Bench
2.1
Terminal-Bench 2.1 · source setting 2: Older, easier version of Terminal-Bench; Still widely reported
AA
Claude Fable 5.1
91.4%
156
+1 versions & slices
4.0
Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Claude Sonnet 5.5
63.6%
132
Terminal-Bench Science
0.1
Terminal-Bench Science 0.1: Agents complete expert-authored research workflows across five scientific domains in a terminal
AA
GPT-6 Astra
63.3%
33
AA-Briefcase
1.1 · Elo
AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating
AA
Claude Sonnet 5.5
1,817
140
DeepSWE
1.1
DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition
GPT-6 Astra
74.1%
44
EnterpriseOps-Gym-AA
EnterpriseOps-Gym-AA: Agents complete stateful enterprise workflows in eight domains; final database state determines success
AA
Claude Fable 5
51.1%
43
Furniture Assembly
Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step
Epoch
Claude Opus 5.5
83.3%
31
FrontierCode
1.1
FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements
AA
Claude Opus 5.5
57.1%
28
EBR Bench
EBR Bench: Agents learn to play Earthborne Rangers across repeated playthroughs
Epoch
GPT-6 Astra
76.2%
24
SWE-bench Pro (v2 full)
SWE-bench Pro (v2 full): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Scale
Claude Opus 5
99.4%
16
SWE-bench Pro (v2 hard)
SWE-bench Pro (v2 hard): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Scale
Claude Opus 5
98%
16
SWE-bench Pro (private)
SWE-bench Pro (private): Agents resolve long-horizon professional repository tasks; public and private splits are separate
Scale
Muse Spark 1.1
51.5%
9
SpireClimb
SpireClimb: Agents play Slay the Spire under the named game and difficulty setting
Epoch
GPT-6 Astra
25.7%
7
SpireClimb StS1 A20 Heart
SpireClimb StS1 A20 Heart: Agents play Slay the Spire under the named game and difficulty setting
Epoch
GPT-6.1 Sol
20%
7
SpireClimb StS2 A10
SpireClimb StS2 A10: Agents play Slay the Spire under the named game and difficulty setting
Epoch
GPT-6 Astra
34.7%
7
Health & law
1
Harvey LAB-AA
1.1
Harvey LAB-AA 1.1: Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit
AA
Muse Wonder Full
13.1%
23
Documents
1
AA-LCR (long context)
1.1
AA-LCR (long context) 1.1: Questions that need reading and connecting facts across very long documents (around 100K tokens)
AA
Kimi K3
88.7%
273
Reported by labs
40
GDPVal-AA
v2.1
Reported by the lab with its own setup
Lab
Kimi K3.1 Pro
1,846
1
OSWorld-Verified
Reported by the lab with its own setup
Lab
Claude Opus 4.8
83.4
5
OSWorld V2.1
Reported by the lab with its own setup
Lab
Kimi K3.1 Pro
79.6
1
Toolathlon
Reported by the lab with its own setup
Lab
GPT-5.4
54.6
6
MCP Atlas
Reported by the lab with its own setup
Lab
Muse Spark 1.1
88.1
5
BrowseComp
Reported by the lab with its own setup
Lab
Gemini 3.1 Pro
85.9
7
BrowseComp-zh
Reported by the lab with its own setup
Lab
DeepSeek V3.1 Terminus
45
1
Tau2-Bench
Reported by the lab with its own setup
Lab
Claude Sonnet 4.5
84.7
4
SWE-bench Pro (edition unstated)
Reported by the lab with its own setup
Lab
Claude Opus 5.5
89.9
44
DeepSWE (edition unstated)
Reported by the lab with its own setup
Lab
Claude Fable 5
70
11
Finance Agent
v2
Reported by the lab with its own setup
Lab
Muse Spark 1.1
57.2
11
HLE without tools
Reported by the lab with its own setup
Lab
Gemini 3.1 Pro
44.4
10
HLE with tools
Reported by the lab with its own setup
Lab
DeepSeek V4.1 Flash 1017
63.9
10
Agents Last Exam
Reported by the lab with its own setup
Lab
Kimi K3.1 Pro
43.2
7
Finch FinWorkBench
Reported by the lab with its own setup
Lab
Kimi K3
77.3
7
CyberGym-E2E AA
Reported by the lab with its own setup
Lab
Mistral Large 4
82
7
MCPAtlas Public
Reported by the lab with its own setup
Lab
Claude Opus 4.6
73.8
6
Automation-Bench
Reported by the lab with its own setup
Lab
DeepSeek V4.1 Flash 1017
54.8
6
Harvey Legal Agent Benchmark
Reported by the lab with its own setup
Lab
Mistral Large 4
15.8
6
B3 Attack Resistance
Reported by the lab with its own setup
Lab
Mistral Large 4
93.3
6
Harmful cyber request refusal mean
Reported by the lab with its own setup
Lab
Mistral Large 4
95.3
6
AA Briefcase combined Elo
Reported by the lab with its own setup
Lab
MiMo v2.6 Pro
1,516
6
KORA aggregate
Reported by the lab with its own setup
Lab
Mistral Large 4
1.7
6
JobBench
Reported by the lab with its own setup
Lab
Muse Spark 1.1
54.7
5
Humanity’s Last Exam with tools
Reported by the lab with its own setup
Lab
Muse Spark 1.1
62.1
5
DeepSearchQA
Reported by the lab with its own setup
Lab
GPT-5.5
87.8
5
CyberGym
Reported by the lab with its own setup
Lab
DeepSeek V4.1 Flash 1017
88.1
5
Cybench
Reported by the lab with its own setup
Lab
Mistral Large 4
93
5
MCP-Universe
Reported by the lab with its own setup
Lab
GPT-5
47.9
4
MCP-Mark
Reported by the lab with its own setup
Lab
GPT-5
50.9
4
Tool-Decathlon
Reported by the lab with its own setup
Lab
Claude Sonnet 4.5
38.6
4
Internal consumer agentic eval
Reported by the lab with its own setup
Lab
ChatGPT GPT-6
76.3
2
Consumer agentic eval first_s seconds
Reported by the lab with its own setup
Lab
ChatGPT GPT-5.6
72.1
2
Consumer agentic eval p25_s seconds
Reported by the lab with its own setup
Lab
ChatGPT GPT-5.6
75.2
2
Consumer agentic eval p50_s seconds
Reported by the lab with its own setup
Lab
ChatGPT GPT-5.6
78.3
2
Consumer agentic eval p75_s seconds
Reported by the lab with its own setup
Lab
ChatGPT GPT-6
85.4
2
Consumer agentic eval p100_s seconds
Reported by the lab with its own setup
Lab
ChatGPT GPT-6
94.2
2
SimpleQA search agent
Reported by the lab with its own setup
Lab
DeepSeek V3.1 Terminus
96.8
1
OfficeQA Pro
v2
Reported by the lab with its own setup
Lab
Kimi K3.1 Pro
64.4
1
Frontier Finance
Reported by the lab with its own setup
Lab
Kimi K3.1 Pro
58.6
1
Legacy · saturated or no longer reported
13
τ²-bench
AA
τ²-Bench: Tests service agents that use tools while coordinating with a simulated user
AA
GLM 5.2
99.1%
242
+8 versions & slices
airline · 0.1.3
τ-bench official · 2-airline-0.1.3: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Claude Sonnet 4.5
70%
8
retail · 0.1.3
τ-bench official · 2-retail-0.1.3: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Claude Sonnet 4.5
86.2%
8
overall · 0.1.3
τ-bench official · 2-overall-0.1.3: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Claude Sonnet 4.5
84.7%
7
telecom · 0.1.3
τ-bench official · 2-telecom-0.1.3: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Qwen3 Max Thinking
98.2%
7
airline · 1.0.1
τ-bench official · 2-airline-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Sierra
Claude Opus 4.5
84%
6
overall · 1.0.1
τ-bench official · 2-overall-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Sierra
Qwen3.5 397B A17B
87.9%
6
retail · 1.0.1
τ-bench official · 2-retail-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Sierra
Qwen3.5 397B A17B
84.4%
6
telecom · 1.0.1
τ-bench official · 2-telecom-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Sierra
Qwen3.5 397B A17B
97.8%
6
BFCL
v4
BFCL v4: Tests function calling, multi-turn interactions, search and memory in the v4 suite
Berkeley
Claude Opus 4.5
77.5%
39
+2 versions & slices
v4 · prompt mode
BFCL v4 · BFCL v4 source function-call format · historical run: Tests function calling, multi-turn interactions, search and memory in the v4 suite
Berkeley
Gemini 2.5 Flash
50.9%
16
v4 · native function calls
BFCL v4 · BFCL v4 source function-call format / FC · historical run: Tests function calling, multi-turn interactions, search and memory in the v4 suite
Berkeley
o3 (2025-04-16)
48.6%
2
SWE-bench Pro (public)
SWE-bench Pro (public): Agents resolve long-horizon professional repository tasks; public and private splits are separate
Muse Spark 1.1
61.5%
19
No benchmark matches
Model
BFCL v4 (past run, native function calls)
Per task
Price
Speed
Supports
Context
o3 (2025-04-16)
OpenAI
48.6
unstated
—
$2
/
$8
—
200K
Qwen3 235B A22B Instruct 2507
Alibaba (Qwen)
Open
48
unstated
—
$0.09
/
$0.55
—
262K
Show 539 not measured on BFCL v4 (past run, native function calls)
