The LLM benchmark aggregator.

Agnes 2.5 Pro Theta

Sapiens AI

Strongest in Agents (#24 of 298, on 2 of its 6 boards), weakest in Core abilities (#107 of 201). Above par in 2 of 4 scopes. Among the models it meets almost everywhere, it finishes behind Gemini 4 Carbon and Claude Fable 5 and ahead of Gemini 3.7 Flash and Gemini 3.1 Pro.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Agents58.4−7.9#24/2982/6

Knowledge work63.5−7.2#30/2082/2

Core abilities49.3−16.3#107/2011/4

General intelligence48.9−20.1#100/1811/3

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Agnes 2.5 Pro Theta, left for the other.

Qwen3.6 27Bwon 3 · lost 0

UI-TARS 1.5 7Bwon 3 · lost 0

Smaug Nemotron 3.5won 3 · lost 0

Gemma 3 4B ITwon 3 · lost 0

Command Awon 3 · lost 0

Qwen3.6 Pluswon 3 · lost 0

Kimi K2.7 Codewon 3 · lost 0

Qwen2.5 Vl 72B Instructwon 3 · lost 0

Grok 4.3won 3 · lost 0

Grok Build 0.1won 3 · lost 0

Gemini 2.5 Flash-Litewon 3 · lost 0

Claude Sonnet 5won 3 · lost 0

Claude Sonnet 4.6won 3 · lost 0

Kimi K2.6won 3 · lost 0

Muse Spark 1.1won 3 · lost 0

NVIDIA Nemotron 3.5 Lightning 30B A3Bwon 4 · lost 0

Gemma 4 31Bwon 4 · lost 0

Nemotron 3.5 Super VLwon 4 · lost 0

GPT-5.4 miniwon 4 · lost 0

Muse Spark Openwon 4 · lost 0

Nemotron 3 Nano 4Bwon 4 · lost 0

Mistral Small 4won 4 · lost 0

GPT-5.4 nanowon 4 · lost 0

North Mini Codewon 4 · lost 0

GPT-4o miniwon 4 · lost 0

Mistral Medium 3.5won 4 · lost 0

Granite 4.2 3Bwon 4 · lost 0

Granite 4.2 8Bwon 4 · lost 0

Command A+won 4 · lost 0

Muse Glimmerwon 4 · lost 0

MiniMax M3won 4 · lost 0

GPT-5.6 Lunawon 4 · lost 0

GPT-4.1 nanowon 4 · lost 0

K-EXAONE 2.0won 4 · lost 0

Mistral Large 4won 4 · lost 0

GPT-6 Lunawon 4 · lost 0

Inklingwon 4 · lost 0

Step 5won 4 · lost 0

Gemini 3.5 Flash-Litewon 4 · lost 0

Llama 4 Scout Instructwon 4 · lost 0

Inkling Smallwon 4 · lost 0

Apodex 1.1won 4 · lost 0

Solar Pro 4won 4 · lost 0

GPT OSS 120Bwon 4 · lost 0

Mistral Small 3.2won 4 · lost 0

Gemma 3 27B ITwon 4 · lost 0

Beam 1.0won 4 · lost 0

K2 Horizon 375B A23Bwon 4 · lost 0

NVIDIA Nemotron 3 Ultra 550B A55Bwon 4 · lost 0

JT-4.1 Flash 236B A21Bwon 4 · lost 0

GLM-4.6won 4 · lost 0

Nanbeige4.3 33Bwon 4 · lost 0

GPT-4.1 miniwon 4 · lost 0

Qwen3 14Bwon 4 · lost 0

Qwen3.8 27Bwon 4 · lost 0

GLM-5.3 Flashwon 3 · lost 1

GLM-5.2won 3 · lost 1

DeepSeek V4 Flashwon 3 · lost 1

MiniMax M3.1 Flashwon 3 · lost 1

Gemini 3.8 Flashwon 3 · lost 1

DeepSeek V4 Flash Visionwon 2 · lost 1

GPT-6 Solwon 2 · lost 1

Muse Spark 1.2won 2 · lost 1

Qwen3.7 Maxwon 2 · lost 1

Gemini 3.6 Flashwon 2 · lost 1

Gemini 3.5 Flashwon 2 · lost 1

Gemini 3.1 Prowon 2 · lost 1

Gemini 3.7 Flashwon 2 · lost 1

Claude Haiku 5.5won 2 · lost 2

Mistral Large 3won 2 · lost 2

DeepSeek V3.2won 2 · lost 2

DeepSeek V3won 2 · lost 2

GPT OSS 20Bwon 2 · lost 2

DeepSeek V4 Prowon 2 · lost 2

Qwen3 30B A3B Thinkingwon 2 · lost 2

Qwen3 32Bwon 2 · lost 2

Llama 4 Maverick Instructwon 2 · lost 2

GPT-5.6 Terrawon 2 · lost 2

DeepSeek V3.1 Terminuswon 2 · lost 2

Claude Sonnet 4.5won 2 · lost 2

Tencent Hy4won 2 · lost 2

Grok 4.6won 2 · lost 2

Muse Emberwon 2 · lost 2

Claude Opus 4.7won 2 · lost 2

Qwen3 Next 80B A3B Thinkingwon 2 · lost 2

GPT-5.4won 2 · lost 2

Gemini 3.1 Flash-Litewon 2 · lost 2

Qwen3 235B A22B Thinkingwon 2 · lost 2

DeepSeek R1won 2 · lost 2

Gemini 2.5 Prowon 2 · lost 2

GPT-5 miniwon 2 · lost 2

GPT-5won 2 · lost 2

Claude Sonnet 4won 2 · lost 2

Qwen3.5 397B A17Bwon 2 · lost 2

Kimi K2.5won 2 · lost 2

Grok 4.5won 2 · lost 2

Claude Opus 4.8won 2 · lost 2

GPT-5.5won 2 · lost 2

Kimi K2won 1 · lost 2

GPT-4.1won 1 · lost 2

Qwen3 235B A22B Instructwon 1 · lost 2

Gemini 2.5 Flashwon 1 · lost 2

GPT-5.2won 1 · lost 2

GPT-5 nanowon 1 · lost 2

O4 Miniwon 1 · lost 2

O3won 1 · lost 2

Grok 4won 1 · lost 2

MiMo V2.6 Prowon 1 · lost 3

Qwen3.8 Maxwon 1 · lost 3

DeepSeek V4.1 Flashwon 1 · lost 3

Kimi K3won 1 · lost 3

Grok 4.7won 1 · lost 3

Muse Spark 1.3won 1 · lost 3

GPT-5.6 Solwon 1 · lost 3

GPT-6.1 Terrawon 1 · lost 3

Claude Opus 5won 0 · lost 3

Claude Opus 4.5won 0 · lost 3

Gemini 3 Prowon 0 · lost 3

MiMo V2.6 Flashwon 0 · lost 4

Ling 3.1 Flashwon 0 · lost 4

Inkling 2 Smallwon 0 · lost 4

GLM-5.3won 0 · lost 4

Qwen3.8 Flash Nextwon 0 · lost 4

Claude Sonnet 5.5won 0 · lost 4

Kimi K3.1 Prowon 0 · lost 4

Muse Wonderwon 0 · lost 4

Muse Wonder Fullwon 0 · lost 4

Claude Opus 5.5won 0 · lost 4

Claude Fable 5.1won 0 · lost 4

Gemini 4 Argonwon 0 · lost 4

GPT-6.1 Solwon 0 · lost 4

Claude Fable 5.2won 0 · lost 4

GPT-6 Astrawon 0 · lost 4

Claude Fable 5won 0 · lost 4

Gemini 4 Carbonwon 0 · lost 4

§ 3 · Sources

Where the numbers come from

3 publications, 3 figures. Every one links to the page it was read from.

Artificial Analysis ↗1 measureread 2026-11-12

Artificial Analysis Intelligence Index 24

GDPval-AA ↗1 measureread 2026-11-12

GDPval-AA 1507

AA-Briefcase ↗1 measureread 2026-11-12

AA-Briefcase 1496

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=agnes-2-5-pro-theta)](https://publicai.io/model-index/m/agnes-2-5-pro-theta)