Best AI Models for Data Analysis

AI models ranked by data analysis capability using MMLU, MATH-500, and GPQA benchmark scores. Find the top LLM for data science, analytics, and insights.

Last updated: 51m ago

#1 Model

Score: 85.7

Average Score

66.2

Across all ranked models

Models Ranked

With benchmark data

Weights:MMLU (40%)MATH-500 (30%)GPQA (30%)

Top Best for Data Analysis Models by Weighted Score

Top 15 models by weighted score

LMMarketCap.com

Benchmark Breakdown

Per-benchmark scores for top 10 models

MMLU

MATH-500

GPQA

LMMarketCap.com

#	Model	Provider	Score	MMLU	MATH-500	GPQA
1	o3OpenAI	OpenAI	85.7	92.3	99	--
2	GPT-5.4OpenAI	OpenAI	85.2	94	95.5	--
3	R1 0528DeepSeek	DeepSeek	84.8	91.5	97.8	--
4	GPT-5.2OpenAI	OpenAI	84.3	93.5	94	--
5	GPT-5.1OpenAI	OpenAI	84	93.2	93.5	--
6	GPT-5OpenAI	OpenAI	83.5	93	92.5	--
7	Gemini 2.5 ProGoogle	Google	83.4	90.8	95.2	--
8	Claude Opus 4.6Anthropic	Anthropic	82.3	92.1	90.5	--
9	DeepSeek V3 0324DeepSeek	DeepSeek	81.4	89.2	92	--
10	Claude Opus 4.5Anthropic	Anthropic	81	91.4	88.1	--
11	DeepSeek V3DeepSeek	DeepSeek	80.3	88.5	90.2	--
12	Claude Sonnet 4.6Anthropic	Anthropic	79.8	91.2	85.3	--
13	Gemini 3 Flash PreviewGoogle	Google	79.2	88	88	--
14	Claude Sonnet 4.5Anthropic	Anthropic	78.7	90.8	83	--
15	Gemini 2.5 FlashGoogle	Google	77.2	85.8	85.8	--
16	Llama 4 MaverickMeta	Meta	76.5	88	81	--
17	GPT-4.1OpenAI	OpenAI	76.2	89.2	78.5	--
18	GPT-4oOpenAI	OpenAI	75.2	88.7	76.6	--
19	Llama 3.3 70B InstructMeta	Meta	74.1	86.3	77	--
20	Gemini 3.1 Pro PreviewGoogle	Google	74.1	92.6	--	--
21	GPT-5.5OpenAI	OpenAI	73.9	92.4	--	--
22	GPT-5.5 ProOpenAI	OpenAI	73.9	92.4	--	--
23	Mistral LargeMistral AI	Mistral AI	72.9	84.7	76	--
24	Claude Haiku 4.5Anthropic	Anthropic	71.4	84.5	72.5	--
25	DeepSeek V3.2DeepSeek	DeepSeek	70.8	88.5	--	--
26	Llama 3.1 70B InstructMeta	Meta	70.5	86	68	--
27	GPT-4o-miniOpenAI	OpenAI	69.2	82	70.2	--
28	Phi 4Microsoft	Microsoft	64.3	84.8	80.4	20.8
29	Llama 4 ScoutMeta	Meta	60.3	79.6	50.3	--
30	Gemma 2 27BGoogle	Google	60.2	75.2	--	--
31	Command R7B (12-2024)Cohere	Cohere	6.3	--	--	7.8
32	Llama 3.1 8B InstructMeta	Meta	5.9	--	--	7.4
33	Llama 3.2 3B InstructMeta	Meta	4.9	--	--	6.2
34	Qwen2.5 7B InstructAlibaba	Alibaba	4.4	--	--	5.5
35	Crouton Alphastealth	stealth	1.6	--	--	2

How scores are calculated

Each model's score is a weighted average of its available benchmark results. When a model is missing some benchmarks, the weights are re-normalized across the benchmarks that are available. All scores are on a 0-100 scale. Data sourced from official model cards, published papers, and third-party evaluation platforms.

Other Specialty Leaderboards

Best for Coding Best for Math Best for Reasoning Best for Writing Best for Instructions Best for Roleplay Best for Multilingual

Frequently Asked Questions

Based on our benchmark analysis, o3 by OpenAI is currently the #1 ranked model for data analysis, with a weighted score of 85.7/100.

Models are ranked using a weighted average of MMLU, MATH-500, GPQA benchmark scores. All scores are normalized to a 0-100 scale.

We currently rank 35 models that have relevant benchmark data for data analysis tasks.