The LLM benchmark aggregator.

Gemini 1.5 Pro

Google

Strongest in Multimodal understanding (#3 of 19), weakest in Secure code (#252 of 274). Above par in 8 of 20 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 4.5 and GPT-5.1 and ahead of Command A and GPT-4.1 Nano.

§ 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.

Knowledge54.7−8.8#58/1382/2

Multimodal understanding66.9−5.2#3/191/1

Academic knowledge42.6−21.3#103/1231/1

Professional41.8−19.8#151/1681/1

Medical46.3−20.5#103/1401/1

Legal43−24.1#118/1511/1

Finance37.6−25.3#135/1511/1

Reasoning37.9−29#161/1783/4

Science37.6−25.9#103/1221/1

Mathematics37.4−28.3#118/1401/2

Reasoning38.4−30.7#120/1402/3

Human preference52.3−15.2#165/3421/1

Safety49.9−11.6#181/3372/3

Safe-prompt compliance41.7−18.7#71/821/1

Harm refusal53.7−7.9#95/3002/2

Jailbreak resistance56.1−8.8#100/2721/1

Fairness50.1−19.9#115/3002/2

Toxicity avoidance52.4−6.6#170/2721/1

Secure code30.2−36.3#252/2741/1

§ 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 Gemini 1.5 Pro, left for the other.

Jamba 1.5 Miniwon 11 · lost 1

Jamba 1.6 Miniwon 13 · lost 2

Jamba 1.6 Largewon 13 · lost 2

Jamba 1.5 Largewon 10 · lost 2

Command R Pluswon 14 · lost 3

Gemini 1.5 Flashwon 14 · lost 3

Mistral Smallwon 13 · lost 3

Mistral Largewon 11 · lost 5

Llama 3.3 Instruct Turbo (70B)won 9 · lost 5

GPT-4o Miniwon 12 · lost 7

Llama 4 Scout Instructwon 12 · lost 7

GPT-4.1 Nanowon 11 · lost 8

Command Awon 9 · lost 8

Ling 3.0 Flashwon 7 · lost 7

GPT OSS 20Bwon 8 · lost 10

Claude 3.5 Haikuwon 7 · lost 9

Grok 4.1 Fastwon 7 · lost 11

Grok 4.3won 7 · lost 11

Llama 4 Maverick Instructwon 7 · lost 12

Gemma 4 31Bwon 5 · lost 9

GPT-4owon 7 · lost 13

Grok 3won 6 · lost 12

MiniMax M3won 6 · lost 12

Mimo V2.6 Flashwon 4 · lost 9

GLM-4.6won 5 · lost 12

DeepSeek R1won 5 · lost 13

DeepSeek V3.2won 5 · lost 13

DeepSeek V4 Flashwon 5 · lost 13

GPT-5 Nanowon 5 · lost 14

GPT OSS 120Bwon 5 · lost 14

GPT-4.1 Miniwon 5 · lost 14

DeepSeek V3won 5 · lost 14

Grok 4won 5 · lost 14

Qwen3.5 Flashwon 3 · lost 9

GPT-5.4 Miniwon 3 · lost 9

GLM-4.5won 4 · lost 13

Mimo V2.5 Prowon 4 · lost 13

O1 Miniwon 3 · lost 10

Mistral Large 3won 3 · lost 10

Kimi K2.5won 4 · lost 14

Kimi K2.6won 4 · lost 14

DeepSeek V4 Prowon 4 · lost 14

GLM-5.2won 4 · lost 14

Gemini 3.5 Flashwon 4 · lost 14

Qwen3.8 Maxwon 4 · lost 14

Gemini 3.6 Flashwon 4 · lost 14

Gemini 2.0 Flashwon 4 · lost 15

Gemini 2.5 Flashwon 4 · lost 15

Gemini 2.5 Flash Litewon 3 · lost 12

GPT-6 Lunawon 3 · lost 12

GPT-6 Solwon 3 · lost 12

Kimi K2 Thinkingwon 3 · lost 13

Inkling Smallwon 2 · lost 10

Qwen3.8 27Bwon 2 · lost 10

GLM-4.7won 2 · lost 10

Gemini 3.5 Flash Litewon 2 · lost 10

DeepSeek V4.1 Flashwon 2 · lost 10

GPT-5.6 Lunawon 3 · lost 15

Grok 4.5won 3 · lost 15

GLM-5.3 Flashwon 3 · lost 15

Qwen3.7 Maxwon 3 · lost 15

GPT-4.1won 3 · lost 16

Kimi K2won 3 · lost 16

Gemini 2.5 Prowon 3 · lost 16

GPT-5.6 Solwon 3 · lost 16

Claude Opus 5won 3 · lost 16

GPT-5.4 Nanowon 2 · lost 11

Grok 4 Fastwon 2 · lost 11

Mimo V2.6 Prowon 2 · lost 11

GPT-4 Turbowon 2 · lost 12

MiniMax M2.7won 2 · lost 15

GLM-5.1won 2 · lost 16

GPT-5.6 Terrawon 2 · lost 16

GPT-5.5won 2 · lost 16

Gemini 3.8 Flashwon 2 · lost 16

O3 Miniwon 2 · lost 17

Claude 3.7 Sonnetwon 2 · lost 17

GPT-5 Miniwon 2 · lost 17

O4 Miniwon 2 · lost 17

Claude Sonnet 4won 2 · lost 17

Qwen3 235B A22B Instructwon 2 · lost 17

O3won 2 · lost 17

Grok 3 Miniwon 1 · lost 11

MiniMax M2.5won 1 · lost 11

Gemini 3.1 Flash Litewon 1 · lost 11

Qwen3.6 Pluswon 1 · lost 11

Claude Opus 4.1won 1 · lost 12

GLM-5won 1 · lost 12

Gemini 3 Flashwon 1 · lost 12

Claude Opus 4.5won 1 · lost 12

Gemini 3.1 Prowon 1 · lost 12

Kimi K3won 1 · lost 12

Claude Sonnet 4.6won 1 · lost 15

Gemini 3 Prowon 1 · lost 15

Grok 4.6won 1 · lost 17

GPT-5.4won 1 · lost 17

Claude 3.5 Sonnetwon 1 · lost 18

O1won 1 · lost 18

Claude Haiku 4.5won 1 · lost 18

MiniMax M2.1won 0 · lost 12

GLM-5.3won 0 · lost 12

Gemini 3.7 Flashwon 0 · lost 12

Claude Fable 5.1won 0 · lost 12

Claude Fable 5won 0 · lost 12

Grok 4.7won 0 · lost 15

Claude Opus 5.5won 0 · lost 15

Claude Opus 4.6won 0 · lost 16

GPT-5.2won 0 · lost 18

Inklingwon 0 · lost 18

Claude Sonnet 5won 0 · lost 18

Claude Opus 4.8won 0 · lost 18

Muse Spark 1.1won 0 · lost 18

Claude Opus 4.7won 0 · lost 18

Claude Opus 4won 0 · lost 19

GPT-5won 0 · lost 19

GPT-5.1won 0 · lost 19

Claude Sonnet 4.5won 0 · lost 19

§ 3 · Sources

Where the numbers come from

7 publications, 23 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1351

ARC-AGI-2 ↗1 measureread 2026-09-26

ARC-AGI-2 0.8%

MMMU-Pro ↗1 measureread 2026-09-26

MMMU-Pro 46.9%

SimpleBench ↗1 measureread 2026-09-26

SimpleBench 27.1%

Vals.ai ↗8 measuresread 2026-09-26

Vals · LegalBench 69.08%Vals · CorpFin 40.52%Vals · TaxEval 59.48%Vals · MortgageTax 56.76%Vals · MedQA 76.53%Vals · GPQA Diamond 58.33%Vals · MMLU Pro 75.29%Vals · AIME 18.75%

HELM Safety ↗5 measuresread 2026-09-26

HELM Safety · HarmBench 79.9%HELM Safety · SimpleSafetyTests 97.5%HELM Safety · Anthropic Red Team 99.9%HELM Safety · BBQ 94.5%HELM Safety · XSTest 90.4%

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26

Enkrypt · Jailbreak risk 7.6%Enkrypt · Harmful content risk 31.7%Enkrypt · CBRN risk 9.8%Enkrypt · Toxicity risk 4.6%Enkrypt · Bias risk 85.3%Enkrypt · Insecure code risk 73.8%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=gemini-1-5-pro)](https://publicai.io/model-index/m/gemini-1-5-pro)