Benchmark Explorer

Compare 298 AI models across four benchmark categories: reasoning, code generation, language understanding, and tool use. Scores are derived from model capabilities and composite rankings.

Total Models

298

With Reasoning

132

With Vision

130

Avg Score

Benchmark Categories

Top models in each benchmark category, scored on capability support and composite ranking data.

Reasoning

Chain-of-thought reasoning, logic, and mathematical problem solving.

#	Model	Provider	Score
1	GPT-5.4 ProOpenAI	OpenAI	100
2	GPT-5.4OpenAI	OpenAI	100
3	Claude Opus 4.6Anthropic	Anthropic	100
4	Claude Sonnet 4.5Anthropic	Anthropic	100
5	Claude Sonnet 4.6Anthropic	Anthropic	100
6	Grok 4.1 FastxAI	xAI	100
7	Grok 4 FastxAI	xAI	100
8	Gemini 3.1 Pro Preview Custom ToolsGoogle	Google	100
9	Gemini 3.1 Pro PreviewGoogle	Google	100
10	Gemini 3 Pro PreviewGoogle	Google	100
11	Gemini 3 Flash PreviewGoogle	Google	100
12	Gemini 3.1 Flash Lite PreviewGoogle	Google	100
13	Gemini 2.5 Flash Lite Preview 09-2025Google	Google	100
14	Gemini 2.5 ProGoogle	Google	100
15	Gemini 2.5 Pro Preview 05-06Google	Google	100

Code Generation

Code writing, debugging, and software engineering tasks.

#	Model	Provider	Score
1	GPT-5.4 ProOpenAI	OpenAI	88
2	GPT-5.2 ProOpenAI	OpenAI	87
3	GPT-5 ProOpenAI	OpenAI	87
4	Claude Opus 4.1Anthropic	Anthropic	80
5	o3 ProOpenAI	OpenAI	80
6	o1-proOpenAI	OpenAI	77
7	o3 Deep ResearchOpenAI	OpenAI	74
8	GPT-5.4OpenAI	OpenAI	71
9	Claude Opus 4.6Anthropic	Anthropic	71
10	Claude Opus 4.5Anthropic	Anthropic	71
11	Claude Sonnet 4.5Anthropic	Anthropic	70
12	GPT-5.2OpenAI	OpenAI	70
13	Qwen3 VL 30B A3B ThinkingAlibaba	Alibaba	70
14	Qwen3 VL 235B A22B ThinkingAlibaba	Alibaba	70
15	Claude Opus 4Anthropic	Anthropic	70

Language Understanding

Text comprehension, multi-modal input, and information retrieval.

#	Model	Provider	Score
1	GPT-5.4 ProOpenAI	OpenAI	82
2	GPT-5.2 ProOpenAI	OpenAI	81
3	GPT-5 ProOpenAI	OpenAI	81
4	Claude Opus 4.1Anthropic	Anthropic	75
5	o3 ProOpenAI	OpenAI	75
6	Claude Opus 4Anthropic	Anthropic	71
7	o3 Deep ResearchOpenAI	OpenAI	70
8	GPT-5.4OpenAI	OpenAI	67
9	Claude Opus 4.6Anthropic	Anthropic	67
10	Claude Opus 4.5Anthropic	Anthropic	67
11	Claude Sonnet 4.5Anthropic	Anthropic	66
12	Claude Sonnet 4.6Anthropic	Anthropic	66
13	GPT-5.2OpenAI	OpenAI	66
14	Qwen3 VL 30B A3B ThinkingAlibaba	Alibaba	66
15	Qwen3 VL 235B A22B ThinkingAlibaba	Alibaba	66

Tool Use

Function calling, structured output, and agentic workflows.

#	Model	Provider	Score
1	GPT-5.4 ProOpenAI	OpenAI	98
2	GPT-5.2 ProOpenAI	OpenAI	98
3	GPT-5 ProOpenAI	OpenAI	98
4	Claude Opus 4.1Anthropic	Anthropic	98
5	o3 ProOpenAI	OpenAI	98
6	o3 Deep ResearchOpenAI	OpenAI	98
7	GPT-5.4OpenAI	OpenAI	98
8	Claude Opus 4.6Anthropic	Anthropic	98
9	Claude Opus 4.5Anthropic	Anthropic	98
10	Claude Sonnet 4.5Anthropic	Anthropic	98
11	Claude Sonnet 4.6Anthropic	Anthropic	98
12	GPT-5.2OpenAI	OpenAI	98
13	Qwen3 VL 30B A3B ThinkingAlibaba	Alibaba	98
14	Qwen3 VL 235B A22B ThinkingAlibaba	Alibaba	98
15	GPT-5.1OpenAI	OpenAI	98

Cross-Benchmark Comparison

All 298 models ranked by average score across all four benchmark categories.

Model	Provider	Code	Reasoning	Language	Tool Use	Average
GPT-5.4 ProOpenAI	OpenAI	88	100	82	98	92
GPT-5.2 ProOpenAI	OpenAI	87	98	81	98	91
GPT-5 ProOpenAI	OpenAI	87	98	81	98	91
Claude Opus 4.1Anthropic	Anthropic	80	98	75	98	88
o3 ProOpenAI	OpenAI	80	98	75	98	88
o3 Deep ResearchOpenAI	OpenAI	74	98	70	98	85
GPT-5.4OpenAI	OpenAI	71	100	67	98	84
Claude Opus 4.6Anthropic	Anthropic	71	100	67	98	84
Claude Opus 4.5Anthropic	Anthropic	71	98	67	98	84
Claude Sonnet 4.5Anthropic	Anthropic	70	100	66	98	84
Claude Sonnet 4.6Anthropic	Anthropic	69	100	66	98	83
GPT-5.2OpenAI	OpenAI	70	98	66	98	83
Qwen3 VL 30B A3B ThinkingAlibaba	Alibaba	70	98	66	98	83
Qwen3 VL 235B A22B ThinkingAlibaba	Alibaba	70	98	66	98	83
GPT-5.1OpenAI	OpenAI	69	98	65	98	83
GPT-5.3-CodexOpenAI	OpenAI	68	98	65	98	82
GPT-5.2-CodexOpenAI	OpenAI	68	98	65	98	82
GPT-5OpenAI	OpenAI	68	98	65	98	82
Claude Opus 4Anthropic	Anthropic	70	98	71	90	82
GPT-5.1-Codex-MaxOpenAI	OpenAI	68	98	64	98	82
o4 Mini Deep ResearchOpenAI	OpenAI	68	98	64	98	82
Grok 4.1 FastxAI	xAI	66	100	63	98	82
Grok 4 FastxAI	xAI	66	100	63	98	82
Gemini 3.1 Pro Preview Custom ToolsGoogle	Google	70	100	58	98	82
Gemini 3.1 Pro PreviewGoogle	Google	70	100	58	98	82
Gemini 3 Pro PreviewGoogle	Google	70	100	58	98	82
GPT-5 MiniOpenAI	OpenAI	67	98	63	98	82
GPT-5 NanoOpenAI	OpenAI	66	98	63	98	81
Claude Haiku 4.5Anthropic	Anthropic	66	98	62	98	81
Gemini 3 Flash PreviewGoogle	Google	68	100	56	98	81
o3OpenAI	OpenAI	65	98	61	98	81
Gemini 3.1 Flash Lite PreviewGoogle	Google	67	100	56	98	80
Gemini 2.5 Flash Lite Preview 09-2025Google	Google	67	100	56	98	80
Gemini 2.5 ProGoogle	Google	67	100	56	98	80
o4 Mini HighOpenAI	OpenAI	64	98	61	98	80
o4 MiniOpenAI	OpenAI	64	98	61	98	80
Gemini 2.5 Pro Preview 05-06Google	Google	67	100	55	98	80
Gemini 2.5 Flash LiteGoogle	Google	66	100	55	98	80
Gemini 2.5 FlashGoogle	Google	66	100	55	98	80
Gemini 2.5 Pro Preview 06-05Google	Google	66	100	54	98	80
Qwen3.5 Plus 2026-02-15Alibaba	Alibaba	65	100	54	98	79
Qwen3.5-FlashAlibaba	Alibaba	65	100	53	98	79
GPT-5.1-CodexOpenAI	OpenAI	65	98	54	98	79
GPT-5 CodexOpenAI	OpenAI	65	98	54	98	79
Grok 4xAI	xAI	62	95	59	98	79
Seed-2.0-MiniByteDance	ByteDance	64	98	53	98	78
Qwen3.5-35B-A3BAlibaba	Alibaba	64	98	53	98	78
Qwen3.5-27BAlibaba	Alibaba	64	98	53	98	78
Qwen3.5-122B-A10BAlibaba	Alibaba	64	98	53	98	78
Qwen3.5 397B A17BAlibaba	Alibaba	64	98	53	98	78
Seed 1.6 FlashByteDance	ByteDance	63	98	52	98	78
Seed 1.6ByteDance	ByteDance	63	98	52	98	78
GPT-5.1-Codex-MiniOpenAI	OpenAI	63	98	52	98	78
Kimi K2.5Moonshot AI	Moonshot AI	62	98	52	98	78
Qwen3 VL 8B ThinkingAlibaba	Alibaba	61	98	51	98	77
Claude Sonnet 4Anthropic	Anthropic	59	98	61	90	77
Nemotron Nano 12B 2 VL (free)NVIDIA	NVIDIA	61	98	55	90	76
Claude 3.7 SonnetAnthropic	Anthropic	57	98	59	90	76
Claude 3.7 Sonnet (thinking)Anthropic	Anthropic	57	98	59	90	76
Qwen3 235B A22B Thinking 2507Alibaba	Alibaba	60	95	48	98	75
Nova 2 LiteAmazon	Amazon	58	100	52	90	75
Grok Code Fast 1xAI	xAI	60	95	47	98	75
Qwen Plus 0728 (thinking)Alibaba	Alibaba	59	100	38	98	74
GPT-5 Image MiniOpenAI	OpenAI	50	98	48	98	74
GPT-5 ImageOpenAI	OpenAI	50	98	48	98	74
MiniMax M2.5MiniMax	MiniMax	58	98	38	98	73
Qwen3 Max ThinkingAlibaba	Alibaba	58	98	38	98	73
MiMo-V2-FlashXiaomi	Xiaomi	58	98	38	98	73
MiniMax M2MiniMax	MiniMax	58	98	38	98	73
Trinity Miniarcee-ai	arcee-ai	58	98	37	98	73
DeepSeek V3.2DeepSeek	DeepSeek	58	98	37	98	73
DeepSeek V3.2 ExpDeepSeek	DeepSeek	58	98	37	98	73
Tongyi DeepResearch 30B A3BAlibaba	Alibaba	58	98	37	98	73
Mercury 2Inception	Inception	57	98	37	98	73
gpt-oss-safeguard-20bOpenAI	OpenAI	57	98	37	98	73
Trinity Mini (free)arcee-ai	arcee-ai	58	95	38	98	72
Nemotron Nano 9B V2 (free)NVIDIA	NVIDIA	58	95	38	98	72
Grok 3 MinixAI	xAI	53	95	42	98	72
R1 0528DeepSeek	DeepSeek	56	98	36	98	72
Step 3.5 Flash (free)StepFun	StepFun	57	98	41	90	72
DeepSeek V3 0324DeepSeek	DeepSeek	55	98	35	98	72
Grok 3 Mini BetaxAI	xAI	52	95	40	98	71
gpt-oss-120b (free)OpenAI	OpenAI	55	98	39	90	71
gpt-oss-20b (free)OpenAI	OpenAI	55	98	39	90	71
o1-proOpenAI	OpenAI	77	98	64	43	71
Qwen3 235B A22BAlibaba	Alibaba	54	95	34	98	70
Solar Pro 3Upstage	Upstage	53	95	33	98	70
MiniMax M2.1MiniMax	MiniMax	53	95	33	98	70
Nemotron 3 Nano 30B A3BNVIDIA	NVIDIA	53	95	33	98	70
Kimi K2 ThinkingMoonshot AI	Moonshot AI	53	95	33	98	70
Llama 3.3 Nemotron Super 49B V1.5NVIDIA	NVIDIA	53	95	33	98	70
DeepSeek V3.1 Terminus (exacto)DeepSeek	DeepSeek	53	95	33	98	70
DeepSeek V3.1 TerminusDeepSeek	DeepSeek	53	95	33	98	70
Qwen3 Next 80B A3B ThinkingAlibaba	Alibaba	53	95	33	98	70
Nemotron Nano 9B V2NVIDIA	NVIDIA	52	95	33	98	70
DeepSeek V3.1DeepSeek	DeepSeek	55	90	35	98	70
Qwen3 4B (free)Alibaba	Alibaba	55	90	35	98	70
ERNIE 4.5 VL 28B A3BBaidu	Baidu	51	90	46	90	69
gpt-oss-120bOpenAI	OpenAI	52	95	32	98	69
gpt-oss-120b (exacto)OpenAI	OpenAI	52	95	32	98	69
gpt-oss-20bOpenAI	OpenAI	52	95	32	98	69
Qwen3 235B A22B Instruct 2507Alibaba	Alibaba	52	95	32	98	69
Qwen3 30B A3BAlibaba	Alibaba	53	93	33	98	69
Qwen3 14BAlibaba	Alibaba	53	93	33	98	69
Qwen3 32BAlibaba	Alibaba	53	93	33	98	69
Step 3.5 FlashStepFun	StepFun	51	98	36	90	69
QwQ 32BAlibaba	Alibaba	52	93	32	98	69
MiniMax M1MiniMax	MiniMax	49	100	35	90	69
Qwen3 8BAlibaba	Alibaba	52	90	33	98	68
Nemotron 3 Nano 30B A3B (free)NVIDIA	NVIDIA	51	95	36	90	68
Qwen3 30B A3B Thinking 2507Alibaba	Alibaba	51	90	32	98	68
GPT-4.1OpenAI	OpenAI	53	53	60	98	66
GPT-5.3 ChatOpenAI	OpenAI	55	48	62	98	66
GPT-5.2 ChatOpenAI	OpenAI	55	48	62	98	66
Sonar Pro SearchPerplexity	Perplexity	64	95	60	43	66
GPT-5.1 ChatOpenAI	OpenAI	54	48	61	98	65
GPT-4.1 MiniOpenAI	OpenAI	51	53	58	98	65
GPT-4.1 NanoOpenAI	OpenAI	51	53	58	98	65
o1OpenAI	OpenAI	57	48	55	98	65
R1DeepSeek	DeepSeek	44	93	29	90	64
Gemini 2.0 Flash LiteGoogle	Google	49	50	48	98	61
Gemini 2.0 FlashGoogle	Google	49	50	48	98	61
Llama 4 MaverickMeta	Meta	46	53	46	98	61
Qwen3 VL 32B InstructAlibaba	Alibaba	48	48	48	98	61
Qwen3 VL 8B InstructAlibaba	Alibaba	48	48	48	98	61
Qwen3 VL 30B A3B InstructAlibaba	Alibaba	48	48	48	98	61
Gemma 3 27B (free)Google	Google	49	45	49	98	60
Mistral Small 3.2 24BMistral AI	Mistral AI	47	48	47	98	60
Nemotron Nano 12B 2 VLNVIDIA	NVIDIA	55	95	45	43	60
Trinity Large Preview (free)arcee-ai	arcee-ai	48	45	46	98	59
Nova Premier 1.0Amazon	Amazon	44	53	49	90	59
Llama 4 ScoutMeta	Meta	45	48	45	98	59
Gemma 3 27BGoogle	Google	45	48	45	98	59
GPT-4o (2024-11-20)OpenAI	OpenAI	45	48	45	98	59
Qwen3 Coder PlusAlibaba	Alibaba	47	53	37	98	59
Mistral Small 3.1 24B (free)Mistral AI	Mistral AI	46	45	46	98	59
Qwen VL MaxAlibaba	Alibaba	44	48	44	98	59
Mistral Large 3 2512Mistral AI	Mistral AI	45	45	45	98	58
Qwen3 VL 235B A22B InstructAlibaba	Alibaba	45	45	45	98	58
Qwen3 Coder FlashAlibaba	Alibaba	46	53	36	98	58
Qwen Plus 0728Alibaba	Alibaba	46	53	36	98	58
GPT-4o (extended)OpenAI	OpenAI	43	48	43	98	58
Ministral 3 14B 2512Mistral AI	Mistral AI	44	45	44	98	58
Ministral 3 8B 2512Mistral AI	Mistral AI	44	45	44	98	58
Ministral 3 3B 2512Mistral AI	Mistral AI	44	45	44	98	58
DeepSeek V3.2 SpecialeDeepSeek	DeepSeek	55	98	35	43	58
GPT-4o AudioOpenAI	OpenAI	48	48	37	98	58
Mistral Medium 3.1Mistral AI	Mistral AI	44	45	44	98	58
GPT-4o (2024-08-06)OpenAI	OpenAI	42	48	43	98	58
Hunyuan A13B InstructTencent	Tencent	54	98	34	43	57
Qwen3 MaxAlibaba	Alibaba	46	48	36	98	57
Grok 3xAI	xAI	43	45	42	98	57
GPT-4oOpenAI	OpenAI	41	48	41	98	57
GPT-4o-mini (2024-07-18)OpenAI	OpenAI	40	48	41	98	57
GPT-4o-miniOpenAI	OpenAI	40	48	41	98	57
GPT-4 TurboOpenAI	OpenAI	42	45	42	98	57
Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google	Google	50	93	40	43	57
Qwen3 Coder NextAlibaba	Alibaba	45	48	35	98	57
Nano Banana Pro (Gemini 3 Pro Image Preview)Google	Google	50	93	40	43	57
KAT-Coder-Pro V1Kuaishou	Kuaishou	45	48	35	98	57
LongCat Flash ChatMeituan	Meituan	45	48	35	98	57
Qwen3 30B A3B Instruct 2507Alibaba	Alibaba	45	48	35	98	57
Mistral Medium 3Mistral AI	Mistral AI	41	45	42	98	57
Grok 3 BetaxAI	xAI	42	45	41	98	57
Sonar Reasoning ProPerplexity	Perplexity	46	95	50	35	57
Qwen3 Next 80B A3B Instruct (free)Alibaba	Alibaba	46	45	36	98	56
GPT-4o (2024-05-13)OpenAI	OpenAI	41	45	41	98	56
Olmo 3.1 32B ThinkAllen AI	Allen AI	54	93	34	43	56
Olmo 3 32B ThinkAllen AI	Allen AI	54	93	34	43	56
Olmo 3 7B ThinkAllen AI	Allen AI	54	93	34	43	56
Qwen3 Coder 480B A35B (exacto)Alibaba	Alibaba	44	48	34	98	56
Qwen-PlusAlibaba	Alibaba	41	53	32	98	56
Qwen3 Coder 30B A3B InstructAlibaba	Alibaba	43	48	34	98	56
MercuryInception	Inception	43	48	33	98	56
o3 Mini HighOpenAI	OpenAI	43	48	33	98	56
Claude 3.5 HaikuAnthropic	Anthropic	37	45	50	90	56
Jamba Large 1.7AI21 Labs	AI21 Labs	44	45	34	98	55
ERNIE 4.5 VL 424B A47B Baidu	Baidu	49	93	44	35	55
o3 MiniOpenAI	OpenAI	42	48	33	98	55
Pixtral Large 2411Mistral AI	Mistral AI	39	45	39	98	55
R1 Distill Llama 70BDeepSeek	DeepSeek	49	98	30	43	55
Claude 3.5 SonnetAnthropic	Anthropic	40	45	45	90	55
Qwen3 Coder 480B A35B (free)Alibaba	Alibaba	43	48	38	90	55
Mercury CoderInception	Inception	41	48	32	98	55
Cogito v2.1 671Bdeepcogito	deepcogito	50	95	30	43	55
DeepSeek V3DeepSeek	DeepSeek	40	48	30	98	54
Devstral 2 2512Mistral AI	Mistral AI	41	45	31	98	54
Kimi K2 0905 (exacto)Moonshot AI	Moonshot AI	41	45	31	98	54
Qwen3 Next 80B A3B InstructAlibaba	Alibaba	40	45	31	98	54
Kimi K2 0905Moonshot AI	Moonshot AI	40	45	31	98	54
Codestral 2508Mistral AI	Mistral AI	40	45	30	98	53
R1 Distill Qwen 32BDeepSeek	DeepSeek	48	93	29	43	53
Llama 3.3 70B InstructMeta	Meta	38	48	29	98	53
ERNIE 4.5 21B A3B ThinkingBaidu	Baidu	47	98	32	35	53
Qwen3 Coder 480B A35BAlibaba	Alibaba	39	45	30	98	53
Kimi K2 0711Moonshot AI	Moonshot AI	39	45	30	98	53
Aion-2.0aion-labs	aion-labs	46	98	32	35	53
Devstral MediumMistral AI	Mistral AI	39	45	29	98	53
Llama 3.1 Nemotron 70B InstructNVIDIA	NVIDIA	37	48	28	98	53
Devstral Small 1.1Mistral AI	Mistral AI	38	45	29	98	53
Qwen-TurboAlibaba	Alibaba	38	45	29	98	53
Nova Pro 1.0Amazon	Amazon	35	45	40	90	53
GPT-4 (older v0314)OpenAI	OpenAI	41	40	31	98	53
GPT-4OpenAI	OpenAI	41	40	31	98	53
Nova Lite 1.0Amazon	Amazon	34	45	40	90	52
Voxtral Small 24B 2507Mistral AI	Mistral AI	40	40	30	98	52
Sonar Deep ResearchPerplexity	Perplexity	42	95	36	35	52
Mistral Small 3Mistral AI	Mistral AI	38	43	29	98	52
GPT-4 Turbo PreviewOpenAI	OpenAI	37	45	28	98	52
GPT-4 Turbo (older v1106)OpenAI	OpenAI	37	45	28	98	52
Olmo 3.1 32B InstructAllen AI	Allen AI	39	40	30	98	52
Llama 3.3 70B Instruct (free)Meta	Meta	37	48	32	90	52
Mistral NemoMistral AI	Mistral AI	35	48	26	98	52
Rnj 1 Instructessentialai	essentialai	38	40	29	98	51
Aion-1.0aion-labs	aion-labs	43	98	29	35	51
Qwen-Max Alibaba	Alibaba	38	40	29	98	51
LFM2.5-1.2B-Thinking (free)Liquid AI	Liquid AI	47	90	32	35	51
Command R (08-2024)Cohere	Cohere	35	45	26	98	51
Command R+ (08-2024)Cohere	Cohere	35	45	26	98	51
Aion-1.0-Miniaion-labs	aion-labs	42	98	28	35	51
Mistral Large 2411Mistral AI	Mistral AI	34	45	26	98	51
Mistral Large 2407Mistral AI	Mistral AI	34	45	26	98	51
Virtuoso Largearcee-ai	arcee-ai	34	48	30	90	51
Qwen2.5 72B InstructAlibaba	Alibaba	35	43	26	98	51
GPT-5 ChatOpenAI	OpenAI	50	48	58	43	50
SabaMistral AI	Mistral AI	34	40	26	98	50
Llama 3.1 8B InstructMeta	Meta	33	43	24	98	50
Llama 3.1 405B InstructMeta	Meta	31	45	23	98	49
Claude 3 HaikuAnthropic	Anthropic	28	45	34	90	49
Llama 3.1 70B InstructMeta	Meta	31	45	22	98	49
ERNIE 4.5 21B A3BBaidu	Baidu	35	40	30	90	49
Mixtral 8x7B InstructMistral AI	Mistral AI	31	43	22	98	49
Llama 3 8B InstructMeta	Meta	30	43	22	98	48
Mistral LargeMistral AI	Mistral AI	29	45	21	98	48
Qwen2.5 7B InstructAlibaba	Alibaba	31	40	23	98	48
GPT-3.5 Turbo 16kOpenAI	OpenAI	30	40	22	98	48
Nova Micro 1.0Amazon	Amazon	29	45	25	90	47
Mistral Small CreativeMistral AI	Mistral AI	31	40	27	90	47
Mixtral 8x22B InstructMistral AI	Mistral AI	29	40	21	98	47
GPT-3.5 TurboOpenAI	OpenAI	29	40	21	98	47
GPT-3.5 Turbo (older v0613)OpenAI	OpenAI	28	40	20	98	47
Gemma 3 4B (free)Google	Google	46	40	45	43	44
GPT-4o Search PreviewOpenAI	OpenAI	42	48	40	43	43
GPT AudioOpenAI	OpenAI	45	48	35	43	43
Sonar ProPerplexity	Perplexity	39	45	52	35	43
GPT Audio MiniOpenAI	OpenAI	44	48	34	43	42
GPT-4o-mini Search PreviewOpenAI	OpenAI	40	48	38	43	42
Qwen VL PlusAlibaba	Alibaba	40	45	41	43	42
Nano Banana (Gemini 2.5 Flash Image)Google	Google	40	43	40	43	42
Qwen2.5 VL 72B InstructAlibaba	Alibaba	40	43	40	43	42
MiniMax-01MiniMax	MiniMax	36	53	42	35	42
Llama 3.2 11B Vision InstructMeta	Meta	37	48	38	43	42
Llama Guard 4 12BMeta	Meta	38	45	39	43	41
Qwen2.5 VL 32B InstructAlibaba	Alibaba	37	45	38	43	41
Gemma 3 4BGoogle	Google	37	45	38	43	41
Gemma 3 12BGoogle	Google	37	45	38	43	41
Molmo2 8BAllen AI	Allen AI	38	43	43	35	40
Spotlightarcee-ai	arcee-ai	35	48	41	35	40
Olmo 3 7B InstructAllen AI	Allen AI	41	43	31	43	40
Gemma 3n 2B (free)Google	Google	42	40	32	43	39
Command ACohere	Cohere	39	45	30	43	39
UI-TARS 7B ByteDance	ByteDance	35	45	41	35	39
Gemma 3 12B (free)Google	Google	38	40	43	35	39
Gemma 3n 4B (free)Google	Google	41	40	31	43	39
ERNIE 4.5 300B A47B Baidu	Baidu	39	40	30	43	38
Palmyra X5Writer	Writer	35	50	31	35	38
SonarPerplexity	Perplexity	31	40	45	35	38
Command R7B (12-2024)Cohere	Cohere	34	45	25	43	37
Mistral Small 3.1 24BMistral AI	Mistral AI	29	45	36	35	36
Phi 4Microsoft	Microsoft	34	43	25	43	36
Maestro Reasoningarcee-ai	arcee-ai	31	48	27	35	35
LFM2.5-1.2B-Instruct (free)Liquid AI	Liquid AI	34	40	30	35	35
Qwen2.5 Coder 7B InstructAlibaba	Alibaba	32	40	24	43	35
Granite 4.0 MicroIBM	IBM	29	45	25	35	34
MiniMax M2-herMiniMax	MiniMax	31	40	27	35	33
Llama 3.2 3B Instruct (free)Meta	Meta	27	45	24	35	33
Gemma 2 27BGoogle	Google	28	40	20	43	33
Qwen2.5-VL 7B InstructAlibaba	Alibaba	24	40	31	35	33
Llama 3 70B InstructMeta	Meta	27	40	19	43	32
LFM2-24B-A2BLiquid AI	Liquid AI	28	40	24	35	32
LFM2-8B-A1BLiquid AI	Liquid AI	28	40	24	35	32
LFM2-2.6BLiquid AI	Liquid AI	28	40	24	35	32
Olmo 2 32B InstructAllen AI	Allen AI	25	45	22	35	32
GPT-3.5 Turbo InstructOpenAI	OpenAI	25	40	18	43	32
Llama Guard 3 8BMeta	Meta	24	45	21	35	31
Llemma 7beleutherai	eleutherai	26	40	23	35	31
Llama 3.1 405B (base)Meta	Meta	24	43	21	35	31
Gemma 3n 4BGoogle	Google	25	40	22	35	31
Coder Largearcee-ai	arcee-ai	25	40	22	35	31
Qwen2.5 Coder 32B InstructAlibaba	Alibaba	25	40	22	35	31
Inflection 3 PiInflection	Inflection	24	40	21	35	30
Inflection 3 ProductivityInflection	Inflection	24	40	21	35	30
Llama 3.2 3B InstructMeta	Meta	21	40	18	35	29
Llama 3.2 1B InstructMeta	Meta	21	40	18	35	29
WizardLM-2 8x22BMicrosoft	Microsoft	21	40	18	35	29
Gemma 2 9BGoogle	Google	17	40	15	35	27
LlamaGuard 2 8BMeta	Meta	16	40	14	35	26
Mistral 7B Instruct v0.1Mistral AI	Mistral AI	14	40	12	35	25

Explore More

Dive deeper with capability filters, pricing comparisons, or the full leaderboard.

All Explorers Capabilities Benchmarks Overview Leaderboard Compare Models