Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

GPT-6 Sol

OpenAI

Strongest in Data analysis (#3 of 57), weakest in Toxicity avoidance (#224 of 272). Above par in 25 of 27 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of Muse Spark 1.1 and Claude Opus 4.8.

§ 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.

Core abilities60.6#8/204
Data analysis63.1#3/57
General intelligence63.6#7/204
Language61.3#13/57
Instruction following51.2#29/57
Reasoning60.8#8/178
Mathematics62#7/140
Reasoning60.6#16/140
Coding59.9#10/165
Agentic coding60.3#12/157
Code generation59#13/77
Agents62.2#20/268
Knowledge work65.9#18/178
Human preference62.5#52/342
Human preference62.5#52/342
Professional52#60/168
Biology research60#4/16
Medical54.9#38/140
Finance51.6#78/151
Legal47#93/151
Safety52.4#120/337
Factual grounding58.8#28/101
Fairness54#77/300
Secure code57.6#90/274
Harm refusal50.3#151/300
Jailbreak resistance50.7#178/272
Toxicity avoidance46.6#224/272

§ 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 GPT-6 Sol, left for the other.

Grok Build 0.1won 18 · lost 1
Qwen3.6 Pluswon 18 · lost 1
Gemini 3.5 Flash Litewon 18 · lost 1
Mistral Large 3won 16 · lost 1
Grok 4.3won 23 · lost 2
DeepSeek V3won 21 · lost 2
DeepSeek R1won 21 · lost 2
GPT-5.4 Nanowon 19 · lost 2
GLM-4.6won 19 · lost 2
Llama 4 Scout Instructwon 18 · lost 2
Grok 4won 18 · lost 2
Gemma 4 31Bwon 18 · lost 2
Kimi K2.7 Codewon 17 · lost 2
Grok 4.1 Fastwon 17 · lost 2
DeepSeek V4 Flashwon 23 · lost 3
DeepSeek V3.2won 19 · lost 3
Kimi K2.5won 19 · lost 3
Command Awon 17 · lost 3
GPT-5.4 Miniwon 17 · lost 3
GPT OSS 20Bwon 15 · lost 3
Mistral Smallwon 14 · lost 3
GPT OSS 120Bwon 18 · lost 4
GPT-4o Miniwon 17 · lost 4
Kimi K2won 17 · lost 4
MiniMax M3won 20 · lost 5
Gemini 2.0 Flashwon 15 · lost 4
Qwen3.8 27Bwon 15 · lost 4
Grok 3won 14 · lost 4
Gemini 2.5 Flashwon 17 · lost 5
Claude Haiku 4.5won 17 · lost 5
GPT-6 Lunawon 20 · lost 6
GPT-4.1 Nanowon 16 · lost 5
Qwen3 235B A22B Instructwon 16 · lost 5
Mimo V2.5 Prowon 15 · lost 5
Claude Sonnet 4.6won 18 · lost 6
Gemini 2.5 Prowon 17 · lost 6
Kimi K2.6won 19 · lost 7
Qwen3.7 Maxwon 18 · lost 7
GLM-5.2won 18 · lost 7
Claude 3.5 Haikuwon 12 · lost 5
Claude 3.5 Sonnetwon 12 · lost 5
O3 Miniwon 14 · lost 6
Claude Sonnet 4won 16 · lost 7
Gemini 3.6 Flashwon 18 · lost 8
GPT-4owon 13 · lost 6
GPT-5 Nanowon 13 · lost 6
Claude Opus 4.6won 15 · lost 7
MiniMax M2.7won 12 · lost 6
GPT-4.1 Miniwon 14 · lost 7
GPT-5.6 Lunawon 17 · lost 9
DeepSeek V4 Prowon 17 · lost 9
Claude Opus 4.5won 13 · lost 7
Kimi K2 Thinkingwon 11 · lost 6
Gemini 3.5 Flashwon 16 · lost 9
O4 Miniwon 14 · lost 8
Claude Opus 4won 13 · lost 8
Gemini 3.1 Prowon 13 · lost 8
Claude 3.7 Sonnetwon 11 · lost 7
DeepSeek V4.1 Flashwon 12 · lost 8
Muse Spark 1.2won 12 · lost 8
GPT-5.2won 15 · lost 10
Inklingwon 15 · lost 10
GLM-5.3 Flashwon 15 · lost 10
GPT-5 Miniwon 13 · lost 9
GLM-5.1won 11 · lost 8
Gemini 3.7 Flashwon 11 · lost 8
GPT-5.4won 15 · lost 11
GPT-4.1won 12 · lost 9
Claude Sonnet 5won 14 · lost 11
GPT-5.6 Terrawon 14 · lost 11
GPT-5won 12 · lost 10
Gemini 3.8 Flashwon 14 · lost 12
Claude Opus 4.7won 14 · lost 12
Grok 4.5won 13 · lost 12
Claude Opus 4.8won 13 · lost 12
Muse Spark 1.1won 13 · lost 12
Gemini 3 Prowon 9 · lost 9
GPT-5.1won 10 · lost 10
Kimi K3won 10 · lost 10
Claude Sonnet 4.5won 11 · lost 11
Grok 4.7won 13 · lost 13
Qwen3.8 Maxwon 12 · lost 13
O3won 10 · lost 11
Mimo V2.6 Prowon 9 · lost 10
O1won 8 · lost 9
Grok 4.6won 12 · lost 14
GLM-5.3won 8 · lost 11
GPT-5.6 Solwon 9 · lost 18
Muse Spark 1.3won 5 · lost 13
GPT-5.5won 6 · lost 20
Claude Opus 5won 5 · lost 21
GPT-6 Astrawon 3 · lost 19
Claude Opus 5.5won 2 · lost 24
Claude Fable 5.1won 1 · lost 18
Claude Fable 5won 1 · lost 18

§ 3 · Sources

Where the numbers come from

9 publications, 29 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1457
Artificial Analysis ↗1 measureread 2026-09-26
Artificial Analysis Intelligence Index 48
GDPval-AA ↗1 measureread 2026-09-26
GDPval-AA 1487
AA-Briefcase ↗1 measureread 2026-09-26
AA-Briefcase 1483
LiveBench ↗8 measuresread 2026-09-26
LiveBench 79.3LiveBench · Reasoning 88.7LiveBench · Coding 81.8LiveBench · Agentic Coding 52.9LiveBench · Mathematics 96.4LiveBench · Data Analysis 81.2LiveBench · Language 85.3LiveBench · Instruction Following 68.6
SimpleBench ↗1 measureread 2026-09-26
SimpleBench 73.1%
Vals.ai ↗9 measuresread 2026-09-26
Vals · Legal Research Bench 28.85%Vals · Harvey Legal Agent Benchmark 1.67%Vals · Finance Agent 49.05%Vals · MedCode 47.07%Vals · MedScribe 82.03%Vals · BioMysteryBench 74.81%Vals · Vibe Code Bench 87.82% (OpenHands)Vals · Code Migration 57.2%Vals · ProofBench 83%
Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26
Enkrypt · Jailbreak risk 12.2%Enkrypt · Harmful content risk 0%Enkrypt · CBRN risk 31.8%Enkrypt · Toxicity risk 8.7%Enkrypt · Bias risk 71.6%Enkrypt · Insecure code risk 18.2%
Vectara · Factual consistency 93.5%

Badge

PublicAI Index badge for GPT-6 Sol[![PublicAI Index](https://publicai.io/model-index/badge?model=gpt-6-sol)](https://publicai.io/model-index/m/gpt-6-sol)