Calculator
Price the workload before you build it
Set what one request looks like and how many you expect, and see every model priced against it. The numbers come from the same catalogue the router routes on, at each provider's list price with nothing added.
| Model | In / MTok | Out / MTok | Per request | Per month |
|---|---|---|---|---|
| Claude Haiku 5.5Anthropic | $0.1 | $0.5 | $0.0004 | $4.50 |
| GPT-6 LunaOpenAI | $0.1 | $0.5 | $0.0004 | $4.50 |
| Mistral SmallMistral | $0.15 | $0.6 | $0.0006 | $6.00 |
| DeepSeek V4.1 FlashDeepSeek | $0.15 | $0.6 | $0.0006 | $6.00 |
| GPT-5 miniOpenAI | $0.25 | $2 | $0.0015 | $15.00 |
| Gemini 3.5 Flash-LiteGoogle | $0.3 | $2.5 | $0.0019 | $18.50 |
| Gemini 2.5 FlashGoogle | $0.3 | $2.5 | $0.0019 | $18.50 |
| DeepSeek V4 ProDeepSeek | $0.66 | $1.98 | $0.0023 | $23.10 |
| Mistral Large 4Mistral | $0.7 | $2.1 | $0.0024 | $24.50 |
| Grok Build 0.1xAI | $1 | $2 | $0.0030 | $30.00 |
| Gemini 3.8 FlashGoogle | $0.75 | $3.75 | $0.0034 | $33.75 |
| Grok 4.3xAI | $1.25 | $2.5 | $0.0037 | $37.50 |
| Kimi K2.6Moonshot AI | $0.95 | $4 | $0.0039 | $39.00 |
| Kimi K2.7 CodeMoonshot AI | $0.95 | $4 | $0.0039 | $39.00 |
| Claude Haiku 4.5Anthropic | $1 | $5 | $0.0045 | $45.00 |
| GPT-6.1 TerraOpenAI | $1 | $5 | $0.0045 | $45.00 |
| Muse Spark 1.3Meta | $1.25 | $4.25 | $0.0046 | $46.25 |
| Muse Spark 1.2Meta | $1.25 | $4.25 | $0.0046 | $46.25 |
| Muse Spark 1.1Meta | $1.25 | $4.25 | $0.0046 | $46.25 |
| Grok 4.7xAI | $2 | $6 | $0.0070 | $70.00 |
| Grok 4.6xAI | $2 | $6 | $0.0070 | $70.00 |
| GPT-5.1OpenAI | $1.25 | $10 | $0.0075 | $75.00 |
| Gemini 2.5 ProGoogle | $1.25 | $10 | $0.0075 | $75.00 |
| Kimi K2.7 Code HighspeedMoonshot AI | $1.9 | $8 | $0.0078 | $78.00 |
| Claude Sonnet 5.5Anthropic | $2 | $10 | $0.0090 | $90.00 |
| Claude Sonnet 5Anthropic | $2 | $10 | $0.0090 | $90.00 |
| GPT-6.1 SolOpenAI | $2 | $10 | $0.0090 | $90.00 |
| Gemini 4 ArgonGoogle | $2 | $10 | $0.0090 | $90.00 |
| Kimi K3.1 ProMoonshot AI | $2 | $10 | $0.0090 | $90.00 |
| Kimi K3Moonshot AI | $3 | $15 | $0.013 | $135.00 |
| Claude Opus 5.5Anthropic | $4 | $20 | $0.018 | $180.00 |
| Muse Wonder 1.0Meta | $4 | $20 | $0.018 | $180.00 |
| Kimi K3.1 Pro HighspeedMoonshot AI | $4 | $20 | $0.018 | $180.00 |
| Claude Opus 5Anthropic | $5 | $25 | $0.022 | $225.00 |
| Claude Fable 5.1Anthropic | $10 | $50 | $0.045 | $450.00 |
| GPT-6 AstraOpenAI | $10 | $50 | $0.045 | $450.00 |
Two things this deliberately leaves out
Caching. Repeated context is billed at a fraction of the price and identical repeats cost nothing at all, so a real bill for agent traffic usually lands under these figures - but by how much depends on your traffic, and a calculator that guessed would be flattering rather than useful.
Routing. These are single-model figures. Sending "auto" spreads a mixed workload across models by what each request actually needs, which is the whole argument for a router - and the saving is measured per request rather than estimated here.
The one number not shown above is our fee, because it is not charged per request: it applies once when you buy credit. The pricing page has the arithmetic.
The same answer, from your code
For anything programmatic - pricing a job before running it, or showing a customer what their request will cost - there is an endpoint that uses the biller's own code rather than a copy of it.
curl https://finalrouter.com/api/v1/cost/estimate \
-H "Authorization: Bearer $FINAL_ROUTER_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "anthropic/claude-sonnet-5-5",
"messages": [{ "role": "user", "content": "..." }],
"max_tokens": 800
}'
# Free, works on an empty balance, and runs the same
# arithmetic the biller runs - so the estimate cannot
# drift away from the invoice.