Benchmarks

Intelligence test

Model quality scored 0–100 by an impartial judge LLM (Claude Sonnet 4.6, blind). Six categories: reasoning, coding, creativity, factual accuracy, instruction-following, and safety.

#1Anthropic
100
Claude Opus 4.7
Tier B
Reasoning95
Coding90
Factual100
#2Anthropic
100
Claude Sonnet 4.6
Tier A
Reasoning95
Coding90
Factual100
#3Anthropic
100
Claude Opus 4.6
Tier B
Reasoning95
Coding90
Factual100
#4Anthropic
100
Claude Opus 4.5
Tier B
Reasoning95
Coding90
Factual100
#5Anthropic
100
Claude Sonnet 4.5
Tier B
Reasoning95
Coding90
Factual100
#6Anthropic
100
Claude Opus 4.1
Tier C
Reasoning95
Coding90
Factual100
#7OpenAI
100
gpt-3.5-turbo
Tier C
Reasoning95
Coding90
Factual100
#8OpenAI
100
gpt-3.5-turbo-1106
Reasoning95
Coding90
Factual100
#9OpenAI
100
gpt-4-turbo
Tier C
Reasoning95
Coding90
Factual100
#10OpenAI
100
gpt-4-turbo-2024-04-09
Tier C
Reasoning95
Coding90
Factual100
#11OpenAI
100
gpt-4o-2024-08-06
Tier C
Reasoning95
Coding90
Factual100
#12OpenAI
100
gpt-4o-2024-11-20
Tier C
Reasoning95
Coding90
Factual100
#13OpenAI
100
gpt-4o-mini-search-preview
Tier C
Reasoning95
Coding90
Factual100
#14OpenAI
100
gpt-4.1
Tier B
Reasoning95
Coding90
Factual100
#15OpenAI
100
gpt-4.1-mini-2025-04-14
Reasoning95
Coding90
Factual100
#16OpenAI
100
gpt-4.1-mini
Tier C
Reasoning95
Coding90
Factual100
#17OpenAI
100
gpt-5-chat-latest
Tier C
Reasoning95
Coding90
Factual100
#18OpenAI
100
gpt-5-search-api
Tier C
Reasoning95
Coding90
Factual100
#19OpenAI
100
gpt-5-search-api-2025-10-14
Reasoning95
Coding90
Factual100
#20OpenAI
100
gpt-4o-search-preview
Tier C
Reasoning95
Coding90
Factual100
#21Google Gemini
100
Gemini Flash-Lite Latest
Tier C
Reasoning95
Coding90
Factual100
#22OVH AI Endpoints (GRA)
100
Qwen2.5-VL-72B-Instruct
Reasoning95
Coding90
Factual100
#23OVH AI Endpoints (GRA)
100
Qwen3-Coder-30B-A3B-Instruct
Reasoning95
Coding90
Factual100
#24Google Gemini
100
Gemini 3.1 Flash Lite
Reasoning95
Coding90
Factual100
#25Anthropic
100
Claude Opus 4.8
Tier A
Reasoning95
Coding90
Factual100
#26OpenAI
99
gpt-4-0613
Reasoning94
Coding89
Factual99
#27OpenAI
99
gpt-4
Tier C
Reasoning94
Coding89
Factual99
#28OpenAI
99
gpt-4o-search-preview-2025-03-11
Reasoning94
Coding89
Factual99
#29OpenAI
98
gpt-4o
Tier C
Reasoning93
Coding88
Factual98
#30OpenAI
98
gpt-4o-2024-05-13
Tier C
Reasoning93
Coding88
Factual98
#31OVH AI Endpoints (GRA)
98
Mistral-Small-3.2-24B-Instruct-2506
Reasoning93
Coding88
Factual98
#32OpenAI
97
gpt-4.1-2025-04-14
Reasoning92
Coding87
Factual97
#33OpenAI
97
gpt-4.1-nano-2025-04-14
Reasoning92
Coding87
Factual97
#34Google Gemini
97
Nano Banana
Reasoning92
Coding87
Factual97
#35OVH AI Endpoints (GRA)
97
Meta-Llama-3_3-70B-Instruct
Reasoning92
Coding87
Factual97
#36Anthropic
96
Claude Haiku 4.5
Tier A
Reasoning91
Coding86
Factual96
#37OpenAI
96
gpt-4o-mini-search-preview-2025-03-11
Reasoning91
Coding86
Factual96
#38Google Gemini
95
Gemini 2.5 Flash-Lite
Tier B
Reasoning90
Coding86
Factual95
#39OpenAI
91
gpt-4.1-nano
Tier C
Reasoning86
Coding82
Factual91
#40Google Gemini
91
Nano Banana 2
Reasoning86
Coding82
Factual91
#41OVH AI Endpoints (GRA)
91
Qwen3-32B
Reasoning86
Coding82
Factual91
#42OpenAI
75
gpt-3.5-turbo-16k
Reasoning71
Coding68
Factual75
#43OpenAI
73
gpt-4o-mini
Tier C
Reasoning69
Coding66
Factual73
#44OpenAI
63
gpt-4o-mini-2024-07-18
Tier C
Reasoning60
Coding57
Factual63
#45OpenAI
53
gpt-3.5-turbo-0125
Reasoning50
Coding48
Factual53
#46OVH AI Endpoints (GRA)
53
Mistral-Nemo-Instruct-2407
Reasoning50
Coding48
Factual53
#47OVH AI Endpoints (GRA)
51
Mistral-7B-Instruct-v0.3
Reasoning48
Coding46
Factual51
#48OVH AI Endpoints (GRA)
48
Llama-3.1-8B-Instruct
Reasoning46
Coding43
Factual48
#49Google Gemini
18
Gemini 3.5 Flash
Tier A
Reasoning17
Coding16
Factual18
#50Google Gemini
5
Gemini 2.5 Flash
Tier A
Reasoning5
Coding5
Factual5
#51Google Gemini
5
Gemini Flash Latest
Tier B
Reasoning5
Coding5
Factual5
#52Google Gemini
5
Gemini 3.1 Pro Preview
Tier C
Reasoning5
Coding5
Factual5
#53Google Gemini
5
Gemini Robotics-ER 1.6 Preview
Reasoning5
Coding5
Factual5
#54Google Gemini
0
Gemini 2.5 Pro
Tier A
Reasoning0
Coding0
Factual0
#55Google Gemini
0
Gemini Pro Latest
Tier C
Reasoning0
Coding0
Factual0
#56Google Gemini
0
Gemini 3 Flash Preview
Tier C
Reasoning0
Coding0
Factual0
#57Google Gemini
0
Gemini 3.1 Pro Preview Custom Tools
Tier C
Reasoning0
Coding0
Factual0
#58Google Gemini
0
Nano Banana Pro
Reasoning0
Coding0
Factual0
#59Google Gemini
0
Nano Banana Pro
Reasoning0
Coding0
Factual0
#60OVH AI Endpoints (GRA)
0
gpt-oss-120b
Tier C
Reasoning0
Coding0
Factual0
#61OVH AI Endpoints (GRA)
0
gpt-oss-20b
Tier C
Reasoning0
Coding0
Factual0
#62OVH AI Endpoints (GRA)
0
Qwen3.5-9B
Reasoning0
Coding0
Factual0
#63OVH AI Endpoints (GRA)
0
Qwen3.5-397B-A17B
Reasoning0
Coding0
Factual0

63 models scored · category breakdown estimated (full per-category scoring in Q3 2026)