API price comparisons · Updated November 2, 2026
LLM API price comparisons
Each comparison puts two models’ API prices side by side: input, cached input, output, Batch and long-prompt rates, scheduled price changes, what a chat reply, document summary, agent task and long document cost on each, and a calculator for your own volumes. The table below shows an agent task of 80,000 input tokens, 60,000 of them cached, and 6,000 output per 1,000 requests.
Claude vs GPT vs Gemini
| Comparison | Input / output per 1M | Agent task per 1,000 |
|---|---|---|
| Claude Sonnet 5.5 vs GPT-6.1 Sol | $2.00 / $10.00 vs $2.00 / $10.00 | $106 vs $106 |
| Claude Opus 5.5 vs GPT-6 Astra | $4.00 / $20.00 vs $10.00 / $50.00 | $212 vs $560 |
| Claude Opus 5.5 vs GPT-6.1 Sol | $4.00 / $20.00 vs $2.00 / $10.00 | $212 vs $106 |
| Claude Fable 5.1 vs GPT-6 Astra | $10.00 / $50.00 vs $10.00 / $50.00 | $515 vs $560 |
| Claude Haiku 5.5 vs GPT-6 Luna | $0.10 / $0.50 vs $0.10 / $0.50 | $5.60 vs $5.60 |
| Claude Sonnet 5.5 vs Gemini 4 Argon | $2.00 / $10.00 vs $2.00 / $10.00 | $106 vs $106 |
| Claude Opus 5.5 vs Gemini 4 Argon | $4.00 / $20.00 vs $2.00 / $10.00 | $212 vs $106 |
| GPT-6.1 Sol vs Gemini 4 Argon | $2.00 / $10.00 vs $2.00 / $10.00 | $106 vs $106 |
| GPT-6 Astra vs Gemini 4 Argon | $10.00 / $50.00 vs $2.00 / $10.00 | $560 vs $106 |
| Claude Sonnet 5.5 vs GPT-6.1 Terra | $2.00 / $10.00 vs $1.00 / $5.00 | $106 vs $53.00 |
| Gemini 3.8 Flash vs GPT-6.1 Terra | $0.75 / $3.75 vs $1.00 / $5.00 | $42.00 vs $53.00 |
| Gemini 3.8 Flash vs Claude Haiku 5.5 | $0.75 / $3.75 vs $0.10 / $0.50 | $42.00 vs $5.60 |
| Gemini 3.8 Flash vs GPT-6 Luna | $0.75 / $3.75 vs $0.10 / $0.50 | $42.00 vs $5.60 |
| Gemini 3.5 Flash-Lite vs Claude Haiku 5.5 | $0.30 / $2.50 vs $0.10 / $0.50 | $22.80 vs $5.60 |
| Gemini 3.5 Flash-Lite vs GPT-6 Luna | $0.30 / $2.50 vs $0.10 / $0.50 | $22.80 vs $5.60 |
| Claude Sonnet 5.5 vs DeepSeek V4.1 Flash | $2.00 / $10.00 vs $0.30 / $1.20 | $106 vs $13.56 |
| GPT-6.1 Sol vs DeepSeek V4.1 Flash | $2.00 / $10.00 vs $0.30 / $1.20 | $106 vs $13.56 |
Claude models compared
| Comparison | Input / output per 1M | Agent task per 1,000 |
|---|---|---|
| Claude Opus 5.5 vs Claude Sonnet 5.5 | $4.00 / $20.00 vs $2.00 / $10.00 | $212 vs $106 |
| Claude Sonnet 5.5 vs Claude Haiku 5.5 | $2.00 / $10.00 vs $0.10 / $0.50 | $106 vs $5.60 |
| Claude Fable 5.1 vs Claude Opus 5.5 | $10.00 / $50.00 vs $4.00 / $20.00 | $515 vs $212 |
OpenAI models compared
| Comparison | Input / output per 1M | Agent task per 1,000 |
|---|---|---|
| GPT-6.1 Sol vs GPT-6 Astra | $2.00 / $10.00 vs $10.00 / $50.00 | $106 vs $560 |
| GPT-6.1 Sol vs GPT-6.1 Terra | $2.00 / $10.00 vs $1.00 / $5.00 | $106 vs $53.00 |
| GPT-6.1 Terra vs GPT-6 Luna | $1.00 / $5.00 vs $0.10 / $0.50 | $53.00 vs $5.60 |
| GPT-6.1 Sol vs GPT-5.6 Sol | $2.00 / $10.00 vs $4.00 / $20.00 | $106 vs $224 |
Gemini models compared
| Comparison | Input / output per 1M | Agent task per 1,000 |
|---|---|---|
| Gemini 4 Argon vs Gemini 3.8 Flash | $2.00 / $10.00 vs $0.75 / $3.75 | $106 vs $42.00 |
| Gemini 3.8 Flash vs Gemini 3.5 Flash-Lite | $0.75 / $3.75 vs $0.30 / $2.50 | $42.00 vs $22.80 |
| Gemini 4 Argon vs Gemini 3.1 Pro Preview | $2.00 / $10.00 vs $2.00 / $12.00 | $106 vs $124 |
Full price lists: Claude API pricing · OpenAI API pricing · Gemini API pricing · all supported models · all prices as JSON or CSV.
The cheaper model per token isn’t always cheaper per customer.
Ganivra prices every call you make on each model and adds it up by customer, feature and workflow, next to what each customer pays.