Compare AI Models

Sort by any column. Click a model name to see full details and pricing.

Model TierContext Input / 1M Output / 1MMMLU MultimodalLink
OpenAI o1
OpenAI
frontier200K$15.00$60.0092.3YesTry
Claude Opus 4.7
Anthropic
frontier200K$15.00$75.0091.8YesTry
DeepSeek R1
DeepSeek
frontier128K$0.550$2.1990.8NoTry
Claude Sonnet 4.6
Anthropic
frontier200K$3.00$15.0088.7YesTry
GPT-4o
OpenAI
frontier128K$2.50$10.0088.7YesTry
Llama 3.1 405B
Meta
frontier128K$3.00$3.0088.6NoTry
DeepSeek V3
DeepSeek
frontier128K$0.270$1.1088.5NoTry
Grok 2
xAI
frontier131K$2.00$10.0087.5NoTry
Perplexity Pro
Perplexity
frontier127K$3.00$15.0087YesTry
Gemini 1.5 Pro
Google
frontier2.0M$1.25$5.0085.9YesTry
Mistral Large 2
Mistral
frontier128K$3.00$9.0084NoTry
GPT-4o mini
OpenAI
budget128K$0.150$0.60082YesTry
Gemini 1.5 Flash
Google
budget1.0M$0.075$0.30078.9YesTry
Gemini 2.0 Flash
Google
budget1.0M$0.100$0.40076YesTry
Claude Haiku 4.5
Anthropic
budget200K$0.800$4.0073.8YesTry

Head-to-Head Comparisons

In-depth breakdowns across 10 real-world categories, with a clear winner for each.

Quick Picks

Best Overall
Claude Sonnet 4.6

Top benchmark scores with excellent coding and instruction following. Strong choice for most tasks.

Best Budget
Gemini 1.5 Flash

Incredibly cheap at $0.075/1M input tokens with a massive 1M context window. Perfect for high-volume tasks.

Best Open Source
Llama 3.1 405B

Self-hostable and competitive with frontier models. Best for privacy-sensitive workloads.