When should I NOT use a reasoning model?

For anything that doesn’t need step-by-step thinking — summaries, simple Q&A, classification, chit-chat. Reasoning models will burn 10× the tokens (and time) for no benefit.

What is the “intelligence index”?

A 0–100 composite of public benchmarks (MMLU, MMLU Pro, GPQA, MATH, HumanEval). Higher is better.

How often is this updated?

Monthly. Each model has a lastUpdated stamp. Open an issue if you spot something stale.

AI Models · Reasoning

Best Reasoning AI Models Reasoning

Long chain-of-thought models built for hard math, code, and planning.

Models

Providers

Reasoning models

7 models matched. Click any column to sort.


GPT-5.5	OpenAI	82	95 tok/s	0.42s	400k	$5.00	$15.00	$7.50
Claude 4 Opus	Anthropic	81	50 tok/s	1.4s	500k	$8.00	$40.00	$16.00
Gemini 2.5 Pro	Google	78	110 tok/s	0.7s	2M	$1.25	$5.00	$2.19
OpenAI o1	OpenAI	76	32 tok/s	12s	200k	$15.00	$60.00	$26.25
Grok 3	xAI	74	75 tok/s	0.6s	1M	$3.00	$15.00	$6.00
DeepSeek R1	DeepSeek	73	60 tok/s	1.5s	128k	$0.55	$2.19	$0.96
OpenAI o3-mini	OpenAI	70	60 tok/s	6.5s	200k	$1.10	$4.40	$1.93

Showing 7 of 7 models. Click any column header to sort. Prices are USD per 1M tokens unless noted otherwise. Estimates marked with *.

Browse by category

Drill into a slice of the catalog — open-source models, video models, or all models from one provider.

By License

Open Source

Models with publicly-released weights you can self-host or fine-tune.

Proprietary

Closed-weights API-only models from major labs.

By Purpose

Frontier

The most capable models from each major lab.

Fast & Cheap

Production-grade workhorses with the best speed and cost.

Code

Models specialised for software engineering tasks.

Multimodal

Models that natively understand text, images, and beyond.

By Modality

Text-to-Image

Generate still images from a text prompt.

Text-to-Video

Generate short video clips from a text prompt.

Text-to-Speech

Generate natural-sounding speech audio from text.

By Provider

OpenAI

All models from OpenAI — GPT, o-series and beyond.

Anthropic

Claude family of models from Anthropic.

Google

Gemini family of models from Google DeepMind.

Mistral

Models from Mistral AI — EU-hosted, multilingual.

DeepSeek

Frontier-class open-weights models from DeepSeek.

xAI

Grok models from xAI.

Frequently asked questions

Reasoning models (o1, o3, DeepSeek R1, Claude 4 Opus) use long internal chain-of-thought before producing an answer. They’re slower and more expensive but dramatically better at hard math, code, and multi-step problems.

Explore the full catalog

See every AI model in one place — intelligence, speed and price on a single sortable table.

All AI models LLM pricing calculator