Mercury 2.5 Preview API Pricing
Mercury 2.5 Preview was the first public version of Inception's Mercury 2.5, a diffusion LLM that generates tokens in parallel rather than one after another. Inception listed it at $0.20/M input, $0.02/M cached input and $0.75/M output but served it only through OpenRouter, never from its own platform. On September 8, 2026 it was replaced by the generally available Mercury 2.5, on Inception's own API at the same list card and with an 80% launch discount on top. This page is kept as the record of the preview.
Run the numbers.
Calculator pre-loaded with the preview's published rates, kept for the record. The generally available Mercury 2.5 has the same list card and is currently 80% off.
Real-world presets.
Rapid coding iteration
Corpus question
Ticket reply
Agent sub-task
This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Mercury 2.5 Preview Current | $0.20 cache $0.02 | $0.75 | $0.108 agentic 92/8 | 260K | Reasoning and long-context work at the lowest Mercury rate |
| Mercury 2.5 | $0.20 cache $0.02 | $0.75 | $0.108 deprecated · same | 260K | The generally available release of this model |
| Mercury 2 | $0.25 cache $0.025 | $0.75 | $0.12 deprecated · +11% | 128K | Production traffic on Inception's own platform |
| Mercury Edit 2 | $0.25 cache $0.025 | $0.75 | $0.12 +11% | 32K | Autocomplete and next-edit |
| Gemini 3.7 Flash | $0.75 cache $0.075 | $3.75 | $0.481 deprecated · +4.4× · promo | 1M | Google's fast mainstream model |
| Claude Haiku 4.5 | $1.00 cache $0.10 | $5.00 | $0.682 +6.3× | 200K | High-volume support and classification |
Frequently asked.
Where you can actually call it, why a preview costs less than the GA model, and what a diffusion LLM changes about the bill.
Q · 01 What happened to Mercury 2.5 Preview? +
mercury-2.5. The developer pricing table now lists no preview row. Inception has not said whether the preview id still answers on OpenRouter, so new work should use the GA model.Q · 02 What did Mercury 2.5 Preview cost? +
$0.20/M input, $0.02/M cached input and $0.75/M output on its models page, about $0.11/M on our 92/8 agentic blend. The GA Mercury 2.5 kept exactly that list card and added an 80% launch discount, so the model became cheaper to use at the moment it left preview.