AI Model Comparison

Compare Leading AI Models

Side-by-side comparison of GPT-4, Claude, and Gemini with performance metrics, pricing, and capabilities

54 models

GPT-5 Thinking Pro

OpenAI

pro / high-end reasoning

Context1000K

Input$8/1M

Output$32/1M

Performance

reasoning100

speed50

accuracy99

creativity98

Capabilities

Max compute reasoning
Parallel test-time compute
Expert-level answers

Aug 2025

GPT-5 Thinking

OpenAI

deep-reasoning / heavy tasks

Context1000K

Input$4/1M

Output$16/1M

Performance

reasoning99

speed60

accuracy98

creativity96

Capabilities

Deep reasoning
Long chain-of-thought
Tool use
Maximal accuracy

Aug 2025

Claude Opus 4.1

Anthropic

flagship / reasoning

Context200K

Input$15/1M

Output$75/1M

Performance

reasoning99

speed60

accuracy98

creativity97

Capabilities

Maximum intelligence
Deep reasoning
Complex tasks
Research

Mar 2025

OpenAI

reasoning

Context200K

Input$2/1M

Output$8/1M

Performance

reasoning99

speed55

accuracy98

creativity94

Capabilities

Deep reasoning
Chain-of-thought
Complex problem solving

Jan 2025

GPT-5 Main

OpenAI

balanced / default

Context1000K

Input$2/1M

Output$8/1M

Performance

reasoning98

speed80

accuracy97

creativity95

Capabilities

Text & vision
Agentic reasoning
Smart routing (fast vs think)
Customizable style / verbosity

Aug 2025

Claude Opus 4

Anthropic

flagship / reasoning

Context200K

Input$15/1M

Output$75/1M

Performance

reasoning98

speed65

accuracy97

creativity95

Capabilities

Text & vision
Coding & reasoning
Tool use
Extended chain-of-thought

May 2025

Claude Sonnet 4.5

Anthropic

flagship / reasoning

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning98

speed80

accuracy97

creativity96

Capabilities

Text & vision
Extended thinking
Tool use
Artifacts

Jan 2025

GPT-5 Thinking Mini

OpenAI

balanced reasoning

Context1000K

Input$2.5/1M

Output$10/1M

Performance

reasoning97

speed75

accuracy96

creativity94

Capabilities

Reasoning with moderate speed
Tool use

Aug 2025

GPT-5 Codex

OpenAI

code-specialized

Context400K

Input$1.25/1M

Output$10/1M

Performance

reasoning97

speed80

accuracy97

creativity92

Capabilities

Code generation
Debugging
Refactoring
Documentation

Aug 2025

Gemini 2.5 Pro

Google

flagship / reasoning

Context1000K

Input$2.5/1M

Output$10/1M

Performance

reasoning97

speed75

accuracy96

creativity95

Capabilities

Text & vision
Deep reasoning
Long context
Multimodal

Sep 2025

GPT-5

OpenAI

balanced / default

Context400K

Input$1.25/1M

Output$10/1M

Performance

reasoning96

speed85

accuracy96

creativity94

Capabilities

Text & vision
Reasoning
Tool use
Multimodal

Aug 2025

Grok 4

xAI

flagship / reasoning

Context256K

Input$3/1M

Output$15/1M

Performance

reasoning96

speed78

accuracy95

creativity96

Capabilities

Real-time data
Deep reasoning
X integration
Humor

Jan 2025

DeepSeek V3.2 Exp Thinking

DeepSeek

reasoning

Context164K

Input$0.28/1M

Output$0.42/1M

Performance

reasoning96

speed70

accuracy94

creativity91

Capabilities

Deep reasoning
Chain-of-thought
Experimental

Mar 2025

Sonar Reasoning Pro

Perplexity

reasoning

Context127K

Input$2/1M

Output$8/1M

Performance

reasoning96

speed72

accuracy94

creativity90

Capabilities

Deep reasoning
Research
Citations
Analysis

Jan 2025

GPT-5 Thinking Nano

OpenAI

fast reasoning nano

Context1000K

Input$1.5/1M

Output$6/1M

Performance

reasoning95

speed90

accuracy94

creativity92

Capabilities

Light reasoning
Low latency

Aug 2025

Claude Sonnet 4

Anthropic

balanced / general

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning95

speed85

accuracy95

creativity92

Capabilities

Text & vision
Balanced reasoning
Tool-enabled
Long context (beta)

May 2025

GPT-4.1

OpenAI

balanced / general

Context1000K

Input$2/1M

Output$8/1M

Performance

reasoning95

speed82

accuracy95

creativity93

Capabilities

Text & vision
Extended context
Tool use

Mar 2025

Sonar Pro

Perplexity

flagship / reasoning

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning95

speed78

accuracy95

creativity91

Capabilities

Advanced search
Deep research
Citations
Analysis

Nov 2024

GPT-5 Main Mini

OpenAI

fast / cost-efficient

Context1000K

Input$1/1M

Output$4/1M

Performance

reasoning94

speed90

accuracy94

creativity90

Capabilities

Text & vision
Low-latency mode
Smart routing

Aug 2025

Claude 3.7 Sonnet

Anthropic

balanced / general

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning94

speed85

accuracy94

creativity91

Capabilities

Text & vision
Tool use
Analysis

Nov 2024

GPT-4o

OpenAI

balanced / general

Context128K

Input$2.5/1M

Output$10/1M

Performance

reasoning94

speed85

accuracy94

creativity92

Capabilities

Text & vision
Audio
Multimodal
Real-time

May 2024

Grok 4 Fast Reasoning

xAI

reasoning

Context2000K

Input$0.2/1M

Output$0.5/1M

Performance

reasoning94

speed88

accuracy93

creativity93

Capabilities

Fast reasoning
Extended context
Real-time

Jan 2025

Claude 3.5 Sonnet

Anthropic

balanced / general

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning93

speed85

accuracy93

creativity90

Capabilities

Text & vision
Tool use
Coding

Jun 2024

Gemini 2.5 Flash

Google

balanced / general

Context1000K

Input$0.3/1M

Output$2.5/1M

Performance

reasoning93

speed90

accuracy93

creativity91

Capabilities

Text & vision
Fast
Long context
Multimodal

Sep 2025

DeepSeek V3.2 Exp

DeepSeek

balanced / general

Context164K

Input$0.27/1M

Output$0.41/1M

Performance

reasoning93

speed85

accuracy92

creativity90

Capabilities

Experimental
Advanced reasoning
Cost-efficient

Mar 2025

GPT-5 Mini

OpenAI

fast / cost-efficient

Context400K

Input$0.25/1M

Output$2/1M

Performance

reasoning92

speed92

accuracy93

creativity90

Capabilities

Text & vision
Fast responses
Cost-efficient

Aug 2025

Grok Code Fast 1

xAI

code-specialized

Context256K

Input$0.2/1M

Output$1.5/1M

Performance

reasoning92

speed90

accuracy92

creativity88

Capabilities

Code generation
Fast
Debugging

Feb 2025

Llama 4 Maverick

Performance

reasoning92

speed88

accuracy91

creativity90

Capabilities

Open-source
Extended context
Customizable

Apr 2025

GPT-4.1 Mini

OpenAI

fast / cost-efficient

Context1000K

Input$0.4/1M

Output$1.6/1M

Performance

reasoning91

speed90

accuracy92

creativity89

Capabilities

Text & vision
Extended context
Fast

Mar 2025

Gemini 2.0 Flash

Google

balanced / general

Context1000K

Input$0.15/1M

Output$0.6/1M

Performance

reasoning91

speed92

accuracy91

creativity89

Capabilities

Text & vision
Fast
Multimodal

Dec 2024

DeepSeek V3.1

DeepSeek

balanced / general

Context164K

Input$0.2/1M

Output$0.8/1M

Performance

reasoning91

speed87

accuracy91

creativity89

Capabilities

Balanced
Cost-efficient
General purpose

Feb 2025

Codestral

Mistral

code-specialized

Context256K

Input$0.3/1M

Output$0.9/1M

Performance

reasoning91

speed88

accuracy92

creativity87

Capabilities

Code generation
80+ languages
Fill-in-middle

May 2024

Nova Pro

Amazon

balanced / general

Context300K

Input$0.8/1M

Output$3.2/1M

Performance

reasoning91

speed86

accuracy91

creativity89

Capabilities

Text & vision
AWS integration
Extended context

Dec 2024

DeepSeek V3

DeepSeek

balanced / general

Context164K

Input$0.77/1M

Output$0.77/1M

Performance

reasoning90

speed86

accuracy90

creativity88

Capabilities

Balanced pricing
General purpose
Reliable

Dec 2024

Sonar

Perplexity

balanced / general

Context127K

Input$1/1M

Output$1/1M

Performance

reasoning90

speed85

accuracy92

creativity88

Capabilities

Web search
Citations
Real-time data

Aug 2024

GLM 4.6

Zhipu AI

balanced / general

Context200K

Input$0.6/1M

Output$2.2/1M

Performance

reasoning90

speed85

accuracy90

creativity88

Capabilities

Chinese & English
Multimodal
Tool use

Oct 2024

GPT-4o Mini

OpenAI

fast / cost-efficient

Context128K

Input$0.15/1M

Output$0.6/1M

Performance

reasoning89

speed92

accuracy90

creativity88

Capabilities

Text & vision
Fast
Affordable

Jul 2024

Qwen3 Coder

Alibaba

code-specialized

Context131K

Input$0.4/1M

Output$1.6/1M

Performance

reasoning89

speed88

accuracy90

creativity86

Capabilities

Code generation
Multiple languages
Debugging

Nov 2024

Claude 3.5 Haiku

Anthropic

fast / low-cost

Context200K

Input$0.8/1M

Output$4/1M

Performance

reasoning88

speed95

accuracy90

creativity87

Capabilities

Fast responses
Cost-efficient
Basic reasoning

Nov 2024

GPT OSS 120B

OpenAI

open-source

Context131K

Input$0.1/1M

Output$0.5/1M

Performance

reasoning88

speed85

accuracy89

creativity87

Capabilities

Open-source
Self-hosted
Customizable

Feb 2025

Gemini 2.5 Flash Lite

Google

fast / low-cost

Context1000K

Input$0.1/1M

Output$0.4/1M

Performance

reasoning88

speed96

accuracy89

creativity87

Capabilities

Ultra-fast
Cost-efficient
Long context

Sep 2025

Kimi K2 0905

Moonshot AI

balanced / general

Context131K

Input$0.6/1M

Output$2.5/1M

Performance

reasoning88

speed84

accuracy89

creativity87

Capabilities

Chinese & English
Long context
General purpose

Sep 2024

GPT-4.1 Nano

OpenAI

fast / low-cost

Context1000K

Input$0.1/1M

Output$0.4/1M

Performance

reasoning87

speed95

accuracy89

creativity86

Capabilities

Extended context
Ultra-fast
Cost-efficient

Mar 2025

Grok 3 Mini

xAI

fast / low-cost

Context131K

Input$0.3/1M

Output$0.5/1M

Performance

reasoning87

speed93

accuracy88

creativity88

Capabilities

Fast
Cost-efficient
Basic reasoning

Nov 2024

Grok 4 Fast Non-Reasoning

xAI

fast / cost-efficient

Context2000K

Input$0.2/1M

Output$0.5/1M

Performance

reasoning86

speed95

accuracy88

creativity90

Capabilities

Ultra-fast
Extended context
Simple tasks

Jan 2025

Claude Haiku

Anthropic

fast / low-cost

Context200K

Input$0.8/1M

Output$4/1M

Performance

reasoning85

speed98

accuracy88

creativity85

Capabilities

Fast responses
Cost-efficient
Basic reasoning

Mar 2024

GPT-5 Nano

OpenAI

fast / low-cost

Context400K

Input$0.05/1M

Output$0.4/1M

Performance

reasoning85

speed98

accuracy88

creativity85

Capabilities

Ultra-fast
Minimal cost
Simple tasks

Aug 2025

Gemini 2.5 Flash Image

Google

image-generation

Context33K

Input$0.3/1M

Output$2.5/1M

Performance

reasoning85

speed80

accuracy88

creativity95

Capabilities

Image generation
Text-to-image
Creative visuals

Sep 2025

Mistral Small

Mistral

fast / low-cost

Context32K

Input$0.1/1M

Output$0.3/1M

Performance

reasoning85

speed93

accuracy86

creativity84

Capabilities

Fast
Cost-efficient
Lightweight

Feb 2024

Gemini 2.0 Flash Lite

Google

fast / low-cost

Context1000K

Input$0.07/1M

Output$0.3/1M

Performance

reasoning84

speed97

accuracy86

creativity84

Capabilities

Ultra-fast
Minimal cost
Basic tasks

Dec 2024

GPT OSS 20B

OpenAI

open-source

Context128K

Input$0.07/1M

Output$0.3/1M

Performance

reasoning82

speed92

accuracy84

creativity83

Capabilities

Open-source
Lightweight
Fast

Feb 2025

Text Embedding 3 Small

OpenAI

embedding

Context8K

Input$0.02/1M

Output$0/1M

Performance

reasoning0

speed98

accuracy90

creativity0

Capabilities

Text embeddings
Semantic search
Clustering

Jan 2024

Text Embedding 3 Large

OpenAI

embedding

Context8K

Input$0.13/1M

Output$0/1M

Performance

reasoning0

speed95

accuracy95

creativity0

Capabilities

High-quality embeddings
Semantic search
RAG

Jan 2024

Gemini Embedding 001

Google

embedding

Context2K

Input$0.15/1M

Output$0/1M

Performance

reasoning0

speed96

accuracy92

creativity0

Capabilities

Text embeddings
Semantic search
Document retrieval

Jun 2024

AI Model Comparison

Compare Leading AI Models

Side-by-side comparison of GPT-4, Claude, and Gemini with performance metrics, pricing, and capabilities

54 models

GPT-5 Thinking Pro

OpenAI

pro / high-end reasoning

Context1000K

Input$8/1M

Output$32/1M

Performance

reasoning100

speed50

accuracy99

creativity98

Capabilities

Max compute reasoning
Parallel test-time compute
Expert-level answers

Aug 2025

GPT-5 Thinking

OpenAI

deep-reasoning / heavy tasks

Context1000K

Input$4/1M

Output$16/1M

Performance

reasoning99

speed60

accuracy98

creativity96

Capabilities

Deep reasoning
Long chain-of-thought
Tool use
Maximal accuracy

Aug 2025

Claude Opus 4.1

Anthropic

flagship / reasoning

Context200K

Input$15/1M

Output$75/1M

Performance

reasoning99

speed60

accuracy98

creativity97

Capabilities

Maximum intelligence
Deep reasoning
Complex tasks
Research

Mar 2025

OpenAI

reasoning

Context200K

Input$2/1M

Output$8/1M

Performance

reasoning99

speed55

accuracy98

creativity94

Capabilities

Deep reasoning
Chain-of-thought
Complex problem solving

Jan 2025

GPT-5 Main

OpenAI

balanced / default

Context1000K

Input$2/1M

Output$8/1M

Performance

reasoning98

speed80

accuracy97

creativity95

Capabilities

Text & vision
Agentic reasoning
Smart routing (fast vs think)
Customizable style / verbosity

Aug 2025

Claude Opus 4

Anthropic

flagship / reasoning

Context200K

Input$15/1M

Output$75/1M

Performance

reasoning98

speed65

accuracy97

creativity95

Capabilities

Text & vision
Coding & reasoning
Tool use
Extended chain-of-thought

May 2025

Claude Sonnet 4.5

Anthropic

flagship / reasoning

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning98

speed80

accuracy97

creativity96

Capabilities

Text & vision
Extended thinking
Tool use
Artifacts

Jan 2025

GPT-5 Thinking Mini

OpenAI

balanced reasoning

Context1000K

Input$2.5/1M

Output$10/1M

Performance

reasoning97

speed75

accuracy96

creativity94

Capabilities

Reasoning with moderate speed
Tool use

Aug 2025

GPT-5 Codex

OpenAI

code-specialized

Context400K

Input$1.25/1M

Output$10/1M

Performance

reasoning97

speed80

accuracy97

creativity92

Capabilities

Code generation
Debugging
Refactoring
Documentation

Aug 2025

Gemini 2.5 Pro

Google

flagship / reasoning

Context1000K

Input$2.5/1M

Output$10/1M

Performance

reasoning97

speed75

accuracy96

creativity95

Capabilities

Text & vision
Deep reasoning
Long context
Multimodal

Sep 2025

GPT-5

OpenAI

balanced / default

Context400K

Input$1.25/1M

Output$10/1M

Performance

reasoning96

speed85

accuracy96

creativity94

Capabilities

Text & vision
Reasoning
Tool use
Multimodal

Aug 2025

Grok 4

xAI

flagship / reasoning

Context256K

Input$3/1M

Output$15/1M

Performance

reasoning96

speed78

accuracy95

creativity96

Capabilities

Real-time data
Deep reasoning
X integration
Humor

Jan 2025

DeepSeek V3.2 Exp Thinking

DeepSeek

reasoning

Context164K

Input$0.28/1M

Output$0.42/1M

Performance

reasoning96

speed70

accuracy94

creativity91

Capabilities

Deep reasoning
Chain-of-thought
Experimental

Mar 2025

Sonar Reasoning Pro

Perplexity

reasoning

Context127K

Input$2/1M

Output$8/1M

Performance

reasoning96

speed72

accuracy94

creativity90

Capabilities

Deep reasoning
Research
Citations
Analysis

Jan 2025

GPT-5 Thinking Nano

OpenAI

fast reasoning nano

Context1000K

Input$1.5/1M

Output$6/1M

Performance

reasoning95

speed90

accuracy94

creativity92

Capabilities

Light reasoning
Low latency

Aug 2025

Claude Sonnet 4

Anthropic

balanced / general

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning95

speed85

accuracy95

creativity92

Capabilities

Text & vision
Balanced reasoning
Tool-enabled
Long context (beta)

May 2025

GPT-4.1

OpenAI

balanced / general

Context1000K

Input$2/1M

Output$8/1M

Performance

reasoning95

speed82

accuracy95

creativity93

Capabilities

Text & vision
Extended context
Tool use

Mar 2025

Sonar Pro

Perplexity

flagship / reasoning

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning95

speed78

accuracy95

creativity91

Capabilities

Advanced search
Deep research
Citations
Analysis

Nov 2024

GPT-5 Main Mini

OpenAI

fast / cost-efficient

Context1000K

Input$1/1M

Output$4/1M

Performance

reasoning94

speed90

accuracy94

creativity90

Capabilities

Text & vision
Low-latency mode
Smart routing

Aug 2025

Claude 3.7 Sonnet

Anthropic

balanced / general

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning94

speed85

accuracy94

creativity91

Capabilities

Text & vision
Tool use
Analysis

Nov 2024

GPT-4o

OpenAI

balanced / general

Context128K

Input$2.5/1M

Output$10/1M

Performance

reasoning94

speed85

accuracy94

creativity92

Capabilities

Text & vision
Audio
Multimodal
Real-time

May 2024

Grok 4 Fast Reasoning

xAI

reasoning

Context2000K

Input$0.2/1M

Output$0.5/1M

Performance

reasoning94

speed88

accuracy93

creativity93

Capabilities

Fast reasoning
Extended context
Real-time

Jan 2025

Claude 3.5 Sonnet

Anthropic

balanced / general

Context200K

Input$3/1M

Output$15/1M

Performance

reasoning93

speed85

accuracy93

creativity90

Capabilities

Text & vision
Tool use
Coding

Jun 2024

Gemini 2.5 Flash

Google

balanced / general

Context1000K

Input$0.3/1M

Output$2.5/1M

Performance

reasoning93

speed90

accuracy93

creativity91

Capabilities

Text & vision
Fast
Long context
Multimodal

Sep 2025

DeepSeek V3.2 Exp

DeepSeek

balanced / general

Context164K

Input$0.27/1M

Output$0.41/1M

Performance

reasoning93

speed85

accuracy92

creativity90

Capabilities

Experimental
Advanced reasoning
Cost-efficient

Mar 2025

GPT-5 Mini

OpenAI

fast / cost-efficient

Context400K

Input$0.25/1M

Output$2/1M

Performance

reasoning92

speed92

accuracy93

creativity90

Capabilities

Text & vision
Fast responses
Cost-efficient

Aug 2025

Grok Code Fast 1

xAI

code-specialized

Context256K

Input$0.2/1M

Output$1.5/1M

Performance

reasoning92

speed90

accuracy92

creativity88

Capabilities

Code generation
Fast
Debugging

Feb 2025

Llama 4 Maverick

Performance

reasoning92

speed88

accuracy91

creativity90

Capabilities

Open-source
Extended context
Customizable

Apr 2025

GPT-4.1 Mini

OpenAI

fast / cost-efficient

Context1000K

Input$0.4/1M

Output$1.6/1M

Performance

reasoning91

speed90

accuracy92

creativity89

Capabilities

Text & vision
Extended context
Fast

Mar 2025

Gemini 2.0 Flash

Google

balanced / general

Context1000K

Input$0.15/1M

Output$0.6/1M

Performance

reasoning91

speed92

accuracy91

creativity89

Capabilities

Text & vision
Fast
Multimodal

Dec 2024

DeepSeek V3.1

DeepSeek

balanced / general

Context164K

Input$0.2/1M

Output$0.8/1M

Performance

reasoning91

speed87

accuracy91

creativity89

Capabilities

Balanced
Cost-efficient
General purpose

Feb 2025

Codestral

Mistral

code-specialized

Context256K

Input$0.3/1M

Output$0.9/1M

Performance

reasoning91

speed88

accuracy92

creativity87

Capabilities

Code generation
80+ languages
Fill-in-middle

May 2024

Nova Pro

Amazon

balanced / general

Context300K

Input$0.8/1M

Output$3.2/1M

Performance

reasoning91

speed86

accuracy91

creativity89

Capabilities

Text & vision
AWS integration
Extended context

Dec 2024

DeepSeek V3

DeepSeek

balanced / general

Context164K

Input$0.77/1M

Output$0.77/1M

Performance

reasoning90

speed86

accuracy90

creativity88

Capabilities

Balanced pricing
General purpose
Reliable

Dec 2024

Sonar

Perplexity

balanced / general

Context127K

Input$1/1M

Output$1/1M

Performance

reasoning90

speed85

accuracy92

creativity88

Capabilities

Web search
Citations
Real-time data

Aug 2024

GLM 4.6

Zhipu AI

balanced / general

Context200K

Input$0.6/1M

Output$2.2/1M

Performance

reasoning90

speed85

accuracy90

creativity88

Capabilities

Chinese & English
Multimodal
Tool use

Oct 2024

GPT-4o Mini

OpenAI

fast / cost-efficient

Context128K

Input$0.15/1M

Output$0.6/1M

Performance

reasoning89

speed92

accuracy90

creativity88

Capabilities

Text & vision
Fast
Affordable

Jul 2024

Qwen3 Coder

Alibaba

code-specialized

Context131K

Input$0.4/1M

Output$1.6/1M

Performance

reasoning89

speed88

accuracy90

creativity86

Capabilities

Code generation
Multiple languages
Debugging

Nov 2024

Claude 3.5 Haiku

Anthropic

fast / low-cost

Context200K

Input$0.8/1M

Output$4/1M

Performance

reasoning88

speed95

accuracy90

creativity87

Capabilities

Fast responses
Cost-efficient
Basic reasoning

Nov 2024

GPT OSS 120B

OpenAI

open-source

Context131K

Input$0.1/1M

Output$0.5/1M

Performance

reasoning88

speed85

accuracy89

creativity87

Capabilities

Open-source
Self-hosted
Customizable

Feb 2025

Gemini 2.5 Flash Lite

Google

fast / low-cost

Context1000K

Input$0.1/1M

Output$0.4/1M

Performance

reasoning88

speed96

accuracy89

creativity87

Capabilities

Ultra-fast
Cost-efficient
Long context

Sep 2025

Kimi K2 0905

Moonshot AI

balanced / general

Context131K

Input$0.6/1M

Output$2.5/1M

Performance

reasoning88

speed84

accuracy89

creativity87

Capabilities

Chinese & English
Long context
General purpose

Sep 2024

GPT-4.1 Nano

OpenAI

fast / low-cost

Context1000K

Input$0.1/1M

Output$0.4/1M

Performance

reasoning87

speed95

accuracy89

creativity86

Capabilities

Extended context
Ultra-fast
Cost-efficient

Mar 2025

Grok 3 Mini

xAI

fast / low-cost

Context131K

Input$0.3/1M

Output$0.5/1M

Performance

reasoning87

speed93

accuracy88

creativity88

Capabilities

Fast
Cost-efficient
Basic reasoning

Nov 2024

Grok 4 Fast Non-Reasoning

xAI

fast / cost-efficient

Context2000K

Input$0.2/1M

Output$0.5/1M

Performance

reasoning86

speed95

accuracy88

creativity90

Capabilities

Ultra-fast
Extended context
Simple tasks

Jan 2025

Claude Haiku

Anthropic

fast / low-cost

Context200K

Input$0.8/1M

Output$4/1M

Performance

reasoning85

speed98

accuracy88

creativity85

Capabilities

Fast responses
Cost-efficient
Basic reasoning

Mar 2024

GPT-5 Nano

OpenAI

fast / low-cost

Context400K

Input$0.05/1M

Output$0.4/1M

Performance

reasoning85

speed98

accuracy88

creativity85

Capabilities

Ultra-fast
Minimal cost
Simple tasks

Aug 2025

Gemini 2.5 Flash Image

Google

image-generation

Context33K

Input$0.3/1M

Output$2.5/1M

Performance

reasoning85

speed80

accuracy88

creativity95

Capabilities

Image generation
Text-to-image
Creative visuals

Sep 2025

Mistral Small

Mistral

fast / low-cost

Context32K

Input$0.1/1M

Output$0.3/1M

Performance

reasoning85

speed93

accuracy86

creativity84

Capabilities

Fast
Cost-efficient
Lightweight

Feb 2024

Gemini 2.0 Flash Lite

Google

fast / low-cost

Context1000K

Input$0.07/1M

Output$0.3/1M

Performance

reasoning84

speed97

accuracy86

creativity84

Capabilities

Ultra-fast
Minimal cost
Basic tasks

Dec 2024

GPT OSS 20B

OpenAI

open-source

Context128K

Input$0.07/1M

Output$0.3/1M

Performance

reasoning82

speed92

accuracy84

creativity83

Capabilities

Open-source
Lightweight
Fast

Feb 2025

Text Embedding 3 Small

OpenAI

embedding

Context8K

Input$0.02/1M

Output$0/1M

Performance

reasoning0

speed98

accuracy90

creativity0

Capabilities

Text embeddings
Semantic search
Clustering

Jan 2024

Text Embedding 3 Large

OpenAI

embedding

Context8K

Input$0.13/1M

Output$0/1M

Performance

reasoning0

speed95

accuracy95

creativity0

Capabilities

High-quality embeddings
Semantic search
RAG

Jan 2024

Gemini Embedding 001

Google

embedding

Context2K

Input$0.15/1M

Output$0/1M

Performance

reasoning0

speed96

accuracy92

creativity0

Capabilities

Text embeddings
Semantic search
Document retrieval

Jun 2024