Claude vs GPT vs Gemini: cost per million tokens compared

The full pricing table (July 2026)

Here is every model from all three providers, sorted by input price per million tokens.

Fast tier ($0.25 to $1.50/MTok input):

• -Gemini 3.5 Flash-Lite: $0.25 input / $1.50 output
• -Gemini 3 Flash: $0.50 / $3.00
• -Claude Haiku 4-5: $1.00 / $5.00
• -GPT 5.6 Luna: $1.00 / $6.00
• -Gemini 3.5 Flash: $1.50 / $9.00

Mid tier ($2.00 to $3.00/MTok input):

• -Gemini 3.1 Pro: $2.00 / $12.00
• -Claude Sonnet 5 (intro pricing): $2.00 / $10.00
• -GPT 5.4: $2.50 / $15.00
• -GPT 5.6 Terra: $2.50 / $15.00
• -Claude Sonnet 4-6: $3.00 / $15.00

Flagship tier ($5.00 to $10.00/MTok input):

• -Claude Opus 4-8: $5.00 / $25.00
• -GPT 5.5: $5.00 / $30.00
• -GPT 5.6 Sol: $5.00 / $30.00
• -Claude Fable 5: $10.00 / $50.00

Cheapest provider by task type

The cheapest model per task type depends on the minimum quality threshold you need. At 80+ confidence score:

• -Classification: Gemini 3.5 Flash-Lite at $0.25/MTok
• -Extraction: Gemini 3.5 Flash-Lite at $0.25/MTok
• -Summarization: Gemini 3 Flash at $0.50/MTok
• -Q&A: Gemini 3 Flash at $0.50/MTok
• -Code generation: Gemini 3.1 Pro at $2.00/MTok
• -Code review: Gemini 3.1 Pro at $2.00/MTok
• -Reasoning: GPT 5.4 at $2.50/MTok
• -Creative writing: GPT 5.4 at $2.50/MTok

Google dominates the budget end of the market. For quality-sensitive workloads, Anthropic and OpenAI compete head-to-head in mid and flagship tiers.

Cache discounts change the math

Anthropic and OpenAI offer 90% off cached input tokens. Google offers 75% off. This matters for workloads with repeated system prompts.

At 80% cache hit rate on a 5,000-token system prompt:

• -Anthropic effective input: $0.50/MTok (for a $5.00 model)
• -OpenAI effective input: $0.50/MTok (for a $5.00 model)
• -Google effective input: $0.50/MTok (for a $2.00 model)

With high cache rates, the gap between providers narrows significantly. Anthropic's deeper discount offsets its higher list price on input-heavy, cache-friendly workloads.

Output pricing matters more than you think

For output-heavy workloads (code generation, creative writing, blog writing), output tokens often outnumber input tokens 2-4x. Output pricing determines total cost:

• -A 2,000-input / 8,000-output call on Opus: $0.21/call
• -Same call on GPT 5.5: $0.25/call (20% more expensive due to $30 output)
• -Same call on Sonnet 5 (intro): $0.084/call (60% cheaper than Opus)

For code generation, Claude Sonnet 5 at introductory pricing is the best value in mid-tier until August 31.

The bottom line

No single provider is cheapest across all workloads. The optimal strategy is routing: send classification to Flash-Lite, code review to Sonnet, and complex reasoning to Opus. A team that routes by task type instead of defaulting to one model saves 40-60% on their total AI bill.

Compare all models for your specific workload at /teardown, or start tracking your multi-provider spend with the free dashboard.