Inception model
Mercury 3 API cost calculator
Mercury 3 costs $0.10 per 1M input tokens and $0.35 per 1M output tokens on the Inception API, with a 1M-token context window. A typical request (2K input, 500 output tokens) costs $0.0004 — about $11.25/month at 1,000 requests per day. Use the calculator below to model your exact workload.
Find me a cheaper modelWhat real workloads cost on Mercury 3
| Workload | Tokens (in / out) | Per request | Per 1K requests |
|---|---|---|---|
| Chatbot message | 500 / 300 | $0.0002 | $0.15 |
| RAG query with context | 4,000 / 500 | $0.0006 | $0.57 |
| Document summarization | 20,000 / 1,000 | $0.0024 | $2.35 |
| Agent coding session | 100,000 / 5,000 | $0.01 | $11.75 |
Mercury 3 vs other Inception models
| Model | Input /1M | Output /1M | Context |
|---|---|---|---|
| Mercury 3 | $0.10 | $0.35 | 1M |
| Mercury 2.5 | $0.04 | $0.15 | 260K |
| Mercury 2 | $0.25 | $0.75 | 128K |
Frequently asked questions
How much does Mercury 3 cost per 1M tokens?
According to current pricing data, Mercury 3 costs $0.10 per 1M input tokens and $0.35 per 1M output tokens. Processing 1M tokens each way costs $0.45.
What does a typical API request to Mercury 3 cost?
A typical request with 2,000 input tokens and 500 output tokens costs $0.0004. At 1,000 requests per day that is $0.38 daily, or about $11.25 per month.
What is Mercury 3's context window?
Mercury 3 supports a 1M-token context window (1,000,000 tokens).
Does Mercury 3 support prompt caching?
Yes. Cached input tokens cost $0.01 per 1M — a 90% discount versus the standard input rate, which matters for agents and chat apps that resend the same system prompt.
What is a cheaper alternative to Mercury 3?
gpt-oss-20b from OpenAI is currently the cheapest comparable option at $0.02 per 1M input tokens versus $0.10 for Mercury 3 — roughly 6x cheaper on input.
Compare all current prices on the live model pricing dashboard, see Mercury 3 vs gpt-oss-20b pricing, or run Mercury 3 head-to-head against other models in the side-by-side comparison playground.
Pricing data via the OpenRouter models API.