11Costs
Pay for thinking, on the record.
Live prices for every model an agent can use, what your agent will cost per month, and the bill as a sum of receipts.
Calculator
What will my agent cost to think?
One decision every
Tokens per decision1,200
Agents1
Claude Sonnet 5.5
$0.32
per month · $0.0053 per decision · 60 decisions
Assumes 70% input and 30% output tokens. Gas and trading fees are separate and appear on each receipt.
Price list
Loading prices…
| Model | In / out · 1M | Context | Per month |
|---|---|---|---|
| Claude Opus 5.5 | $4 / $20 | 1M | $0.63 |
| Claude Sonnet 5.5 | $2 / $10 | 1M | $0.32 |
 GPT-6.1 Sol | $2 / $10 | 1.05M | $0.32 |
 Gemini 4 Argon | $2 / $10 | 1.05M | $0.32 |
 Grok 4.7 | $2 / $6 | 500K | $0.23 |
 DeepSeek V4 Pro | $0.66 / $1.98 | 1.05M | $0.08 |
| GPT-6.1 Terra | $1 / $5 | 1.05M | $0.16 |
 Kimi K2.6 | $0.65 / $3.41 | 262K | $0.11 |
 Mistral Large 4 | $0.7 / $2.1 | 1.05M | $0.08 |
| Gemini 3.8 Flash | $0.825 / $4.125 | 1.05M | $0.13 |
| Claude Haiku 5.5 | $0.1 / $0.5 | 1M | $0.02 |
| GPT-6 Luna Pro | $0.1 / $0.5 | 1.05M | $0.02 |
| DeepSeek V4.1 Flash | $0.045 / $0.5 | 1.05M | $0.01 |
 Qwen3.8 Flash | $0.15 / $0.47 | 1M | $0.02 |
| gpt-oss 120B | $0.15 / $0.6 | 131K | $0.02 |
 Llama 4 Maverick | $0.188 / $0.652 | 524K | $0.02 |
Proof of bill
Model spend, totalled from receipts
Every decision receipt carries the model, its token count and its cost, so the bill is the sum of receipts anyone can verify.
Demo agents simulate their model calls, so this bill is computed at list prices, not charged. Live agents record the real cost OpenRouter reports.