model_comparison
LLM pricing, compared on cost
List prices tell you half the story. These compare Claude, GPT-4o, and Gemini on the bill that actually lands: the monthly cost of a real agent workload, where every tool call re-sends the context. Prices as of 2026-06.
cross_provider
Across providers
Claude against GPT-4o against Gemini, the matchups most teams are actually weighing.
Claude Sonnet vs GPT-4o
GPT-4o is about 23% cheaper on an agent workload ($4,532/mo).
Compare costsClaude Opus vs GPT-4o
GPT-4o is about 85% cheaper on an agent workload ($4,532/mo).
Compare costsClaude Sonnet vs Gemini 1.5 Pro
Gemini 1.5 Pro is about 61% cheaper on an agent workload ($2,267/mo).
Compare costsGPT-4o vs Gemini 1.5 Pro
Gemini 1.5 Pro is about 50% cheaper on an agent workload ($2,267/mo).
Compare costsClaude Opus vs Gemini 1.5 Pro
Gemini 1.5 Pro is about 92% cheaper on an agent workload ($2,267/mo).
Compare costsClaude Haiku vs GPT-4o mini
GPT-4o mini is about 83% cheaper on an agent workload ($274/mo).
Compare costsGPT-4o mini vs Gemini 1.5 Flash
Gemini 1.5 Flash is about 50% cheaper on an agent workload ($138/mo).
Compare costsClaude Haiku vs Gemini 1.5 Flash
Gemini 1.5 Flash is about 91% cheaper on an agent workload ($138/mo).
Compare costsClaude Sonnet vs GPT-4o mini
GPT-4o mini is about 95% cheaper on an agent workload ($274/mo).
Compare costsGPT-4o vs Gemini 1.5 Flash
Gemini 1.5 Flash is about 97% cheaper on an agent workload ($138/mo).
Compare costssame_provider
Tier vs tier, same provider
When to move up or down a tier within Anthropic, OpenAI, or Google.
Claude Haiku vs Claude Sonnet
Claude Haiku is about 73% cheaper on an agent workload ($1,567/mo).
Compare costsClaude Sonnet vs Claude Opus
Claude Sonnet is about 80% cheaper on an agent workload ($5,870/mo).
Compare costsGPT-4o mini vs GPT-4o
GPT-4o mini is about 94% cheaper on an agent workload ($274/mo).
Compare costsGemini 1.5 Flash vs Gemini 1.5 Pro
Gemini 1.5 Flash is about 94% cheaper on an agent workload ($138/mo).
Compare costs