How much does the GPT-6 Luna API cost?

On FastMetal, GPT-6 Luna is billed per token: $0.1055 per 1M input tokens and $0.5275 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.

Can I call GPT-6 Luna with the OpenAI SDK?

Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "gpt-6-luna" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.

What is the context window of GPT-6 Luna?

1,050,000 tokens, shared between the prompt and the response.

What do I need to try GPT-6 Luna?

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.

Back to Models

OpenAI

gpt-6-luna

GPT-6 Luna

GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...

9/22/2026

1,050,000 tokens

Input: $0.1055/M

Output: $0.5275/M

Try in Chat Get API Key

Specifications

Modalities

Input

file

image

text

Output

text

Supported Parameters

include_reasoning

max_completion_tokens

max_tokens

reasoning

reasoning_effort

response_format

seed

structured_outputs

tool_choice

tools

Max Output Tokens

128,000

Reasoning Configuration

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

How to control thinking, and what it costs

Data policy

Prompt retention: Unknown: we could not confirm
Training: Unknown: not declared

"Unknown" does not mean "safe". It means we could not confirm it.

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.

Measured performance

Time to first response (p50)2.6s

Throughput (p50)82 tokens/s

Measured through FastMetal's own gateway over the last 30 days (618 first-response samples).

Throughput is sampled once a day: the median of 25 observations over the last 30 days.

Time to first response is measured to the first streamed chunk; throughput counts all output tokens, including reasoning tokens. These are observations, not a performance guarantee.

Code Examples

curl https://api.fastmetal.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-6-luna",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

How GPT-6 Luna 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.

In **“strawberry”**, the letter **r** appears at positions 3, 8, and 9. So there are **3** “r”s.

2.7s response94 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…

`users` is `undefined` when `UserList` renders, so calling `users.map(...)` throws. This often happens because the parent didn’t pass the prop or because data hasn’t loaded yet. If an empty list is acceptable until the data arrives, provide a default: ```jsx function UserList({ users = [] }) { return ( <ul> {users.map(user => ( <li key={user.id}>{user.name}</li> ))} </ul> ); } ``` Also check that the parent passes `users`: ```jsx <UserList users={users} /> ``` If `users` is loaded asynchronously, initialize it as an array: ```jsx const [users, setUsers] = useState([]); ``` If `null` is also possible, normalize it before mapping, for example `(users ?? […

3.8s response289 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):…

Key improvements: - Move imports to module scope. - Catch `requests.RequestException` instead of using a bare `except`; the current handler can hide programming errors as well as network failures. - Call `raise_for_status()` so unsuccessful HTTP responses are handled explicitly. - Use a list comprehension instead of building the result manually. - Replace `item["active"] == True` with `item.get("active") is True` if the field is expected to be the Boolean `True`. - Add backoff between attempts. A fixed two-second delay can make retries slower than necessary. - Decide whether failures should raise an exception or return `[]`. Returning `[]` makes a failed request indistinguishable from a suc…

16.5s response1713 output tokens

Compare these answers side by side with other models →

Frequently asked questions

How much does the GPT-6 Luna API cost?: On FastMetal, GPT-6 Luna is billed per token: $0.1055 per 1M input tokens and $0.5275 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.
Can I call GPT-6 Luna with the OpenAI SDK?: Yes. Point base_url at https://api.fastmetal.ai/v1 and pass "gpt-6-luna" as the model; existing OpenAI-style code works unchanged, including streaming, tool calls and structured output.
What is the context window of GPT-6 Luna?: 1,050,000 tokens, shared between the prompt and the response.
What do I need to try GPT-6 Luna?: 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.