Back to Models
Inception
mercury-3
Chat only

Mercury 3

Mercury 3 is a diffusion-based language model from Inception, and the lab's most capable model to date. It is suited for agentic workflows and real-time applications where latency compounds, such...

10/16/2026
1,000,000 tokens
Input: $0.11/M
Output: $0.385/M

Specifications

Modalities

Input
text
Output
text

Supported Parameters

include_reasoning
max_tokens
reasoning
reasoning_effort
response_format
stop
structured_outputs
temperature
tool_choice
tools

Max Output Tokens

131,072

Reasoning Configuration

Default
Thinking on (effort: medium)
Selectable effort levels
high
medium
low
none
Turning thinking off
Possible (reasoning.enabled: false)

How to control thinking, and what it costs

Data policy

Prompt retention
Not retained — every host this model runs on is configured not to retain prompts
Training
Not used for training

Whether a provider trains on prompts is a declared value from our terms with them. Retention is determined from the upstream listing for every host this model can reach. Neither is guessed.

Available in the web chat only

Mercury 3 can be used in FastMetal's web chat, billed per token from your prepaid balance like any other model. It is not available through the API or MCP: a request naming it is refused by the gateway (HTTP 403).

Open in chat

How Mercury 3 actually answers

Real responses to our standard prompts, recorded on FastMetal.

Count the number of 'r's in 'strawberry'

Count the number of 'r's in 'strawberry'. Explain your reasoning step by step.

Let’s spell out **“strawberry”** and count each **r**: 1. s 2. t 3. **r** ← 1st r 4. a 5. w 6. b 7. e 8. **r** ← 2nd r 9. **r** ← 3rd r 10. y There are **3** letter **r**’s in **“strawberry”**.
1.1s response232 output tokens

Debug This Error

I'm getting the following error in my Node.js application: TypeError: Cannot read properties of undefined (reading 'map') at UserList (/app/components/UserList.js:12:25) at renderWithHooks (/app/node_modules/rea…

The error happens because `users` is `undefined` when `UserList` renders, and you're trying to call `.map()` on it. In your component: ```js function UserList({ users }) { return ( <ul> {users.map(user => <li key={user.id}>{user.name}</li>)} </ul> ); } ``` If the parent component doesn't pass `users`, or passes it as `undefined`, `users.map` will throw: ``` TypeError: Cannot read properties of undefined (reading 'map') ``` ## How to fix it You have a few good options: ### 1. Provide a default value ```js function UserList({ users = [] }) { return ( <ul> {users.map(user => <li key={user.id}>{user.name}</li>)} </ul> ); } ``` Now if `users` isn't pas…
1.8s response749 output tokens

Code Review

Please review the following Python function and suggest improvements for readability, performance, and best practices: def get_data(url, retries=3): import requests import time for i in range(retries):…

The function works, but it has several issues around imports, error handling, retry logic, readability, and performance. Below is a breakdown and an improved version. ## Key issues 1. **Imports inside function** `import requests` and `import time` should be at the top of the module, not inside the function. 2. **Bare `except`** `except:` catches everything, including `KeyboardInterrupt` and `SystemExit`. Catch specific exceptions like `requests.RequestException` or `ValueError`. 3. **No use of `requests.Session()`** Creating a new connection for every request is inefficient. A session reuses the underlying connection pool. 4. **Manual status code check** `if r.status…
6.1s response2728 output tokens

Compare these answers side by side with other models →

Frequently asked questions

How much does the Mercury 3 API cost?
On FastMetal, Mercury 3 is billed per token: $0.11 per 1M input tokens and $0.385 per 1M output tokens on a US-dollar account, before tax. Usage is drawn from a prepaid balance; there is no subscription or monthly fee.
What is the context window of Mercury 3?
1,000,000 tokens, shared between the prompt and the response.
What do I need to try Mercury 3?
Create an account and add credit; the model is then available both in the browser chat and over the API. There is no contract or minimum spend.

Try Mercury 3 right now

Mercury 3 is available on FastMetal through one API key. Start in the browser, or call it from the OpenAI SDK.