Claude Opus 5 vs GPT-5.6 Luna
Pricing, context window and real answers (August 2026)
Claude Opus 5 (Anthropic) and GPT-5.6 Luna (OpenAI), compared as you can actually call them on FastMetal. Both are served from the same OpenAI-compatible endpoint and API key; switching is a change to the model string.
Specs and pricing
 Claude Opus 5 |  GPT-5.6 Luna | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Input (per 1M tokens) | ¥893.5 | ¥35.74 |
| Output (per 1M tokens) | ¥4,467.5 | ¥214.44 |
| Typical cost (1,000 in + 500 out tokens × 1,000 calls) | ¥3,127 | ¥143 |
| Context window | 1,000,000 tokens | 1,050,000 tokens |
| Release date | 7/24/2026 | 7/9/2026 |
| Input modalities | text, image, file | file, image, text |
| Arena · overall | #7 · ELO 1,492 | #63 · ELO 1,452 |
| Arena · Japanese | #8 · ELO 1,495 | #36 · ELO 1,444 |
| Arena · coding | #8 · ELO 1,533 | #69 · ELO 1,496 |
Prices are FastMetal's yen rates before tax. Arena ranks are as of the latest leaderboard scrape.
Which should you pick?
- On price, GPT-5.6 Luna: ¥214.44 per 1M output tokens, 95% less than Claude Opus 5.
- On the Overall arena board, Claude Opus 5 ranks higher (#7; GPT-5.6 Luna is #63).
- On the Japanese arena board, Claude Opus 5 ranks higher (#8; GPT-5.6 Luna is #36).
- On the Coding arena board, Claude Opus 5 ranks higher (#8; GPT-5.6 Luna is #69).
- For long documents, GPT-5.6 Luna: a 1,050,000-token context window against 1,000,000 for Claude Opus 5.
- If in doubt, try both on the same key. Switching is a change to the model string, and each is billed at its own rate.
Real answers to the same prompts
Exactly what the FastMetal gateway returned, side by side. Not benchmark scores: actual output.
Count the number of 'r's in 'strawberry'
Count the number of 'r's in 'strawberry'. Explain your reasoning step by step.
Claude Opus 5
## Counting the 'r's in "strawberry"
**Step 1: Break the word into individual letters**
| Position | Letter |
|----------|--------|
| 1 | s |
| 2 | t |
| 3 | **r** |
| 4 | a |
| 5 | w |
| 6 | b |
| 7 | e |
| 8 | **r** |
| 9 | **r** |
| 10 | y |
**Step 2: Identify each occurrence of 'r'**
- Position 3: the 'r' in the "str-" beginning
- Position 8: the first 'r' in the "-rry" ending
- Position 9: the second 'r' in the "-rry" ending
**Step 3: Tally them up**
3 + 8 + 9 → that's **3 occurrences**.
## Answer: 3
The tricky part is that "strawberry" contains a *double* 'r' near the end ("berry"), which is easy to skim over as a single letter. Splitting the word into "straw" (1 'r') + "berry"…
7.1s466 output tokens
GPT-5.6 Luna
The word **“strawberry”** is spelled:
**s – t – r – a – w – b – e – r – r – y**
The letter **“r”** appears at positions **3, 8, and 9**.
**Answer: 3**
3.0s120 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…
Claude Opus 5
## What's happening
The error means `users` is `undefined` when `UserList` renders — so calling `.map()` on it throws. React isn't the problem; the prop simply isn't there yet (or ever).
The three most common causes:
1. **The parent isn't passing the prop** (or passes it under a different name / typo):
```jsx
<UserList /> // users is undefined
<UserList user={data} /> // wrong name
```
2. **The data is fetched asynchronously**, and the first render happens before it arrives:
```jsx
const [users, setUsers] = useState(); // undefined on first render!
useEffect(() => { fetch('/api/users').then(r => r.json()).then(setUsers); }, []);
return <UserList user…
15.1s1191 output tokens
GPT-5.6 Luna
`users` is `undefined` when `UserList` renders, so calling `.map()` on it fails. `.map()` only exists on arrays.
Common causes include:
- The parent does not pass a `users` prop.
- The prop name differs, such as `userList` instead of `users`.
- Users are loaded asynchronously and are initially undefined.
- The API response has a different shape, such as `{ data: [...] }`.
### Provide a default value
```jsx
function UserList({ users = [] }) {
return (
<ul>
{users.map(user => (
<li key={user.id}>{user.name}</li>
))}
</ul>
);
}
```
The parent should pass an array:
```jsx
<UserList users={users} />
```
If the list comes from state, initialize it as an empty…
4.8s507 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):…
Claude Opus 5
# Code Review: `get_data`
Overall, the function works for the happy path, but there are several issues worth addressing — the most serious being the bare `except`, the silent failure mode, and the conflation of two responsibilities (fetching and transforming).
## Critical Issues
### 1. Bare `except` swallows everything
```python
except:
time.sleep(2)
```
This catches `KeyboardInterrupt`, `SystemExit`, `MemoryError`, and — importantly — bugs in your own code. If `item["active"]` raises a `KeyError` because the API changed its schema, you'll retry three times and return `[]`, giving you no indication anything went wrong. Catch `requests.RequestException` instead (the base class for all…
34.7s2823 output tokens
GPT-5.6 Luna
### Main issues
- Imports should be at module level, not inside the function.
- A bare `except:` catches everything, including programming errors and `KeyboardInterrupt`.
- The function sleeps even after the final failed attempt.
- It retries all HTTP statuses, including permanent errors such as `404`.
- `status_code == 200` excludes other successful `2xx` responses.
- `if item["active"] == True` should generally be `if item.get("active") is True`.
- A `requests.Session` can reuse connections and improve performance.
- Returning `[]` for every failure makes it impossible to distinguish “no active items” from “request failed.”
- Retries should generally use backoff rather than a fixed delay.…
23.3s1978 output tokens
Frequently asked questions
- Which is cheaper, Claude Opus 5 or GPT-5.6 Luna?
- Per 1M output tokens, Claude Opus 5 is ¥4,467.5 and GPT-5.6 Luna is ¥214.44 on FastMetal (yen, before tax), so GPT-5.6 Luna is cheaper.
- How do the context windows of Claude Opus 5 and GPT-5.6 Luna compare?
- Claude Opus 5 takes 1,000,000 tokens; GPT-5.6 Luna takes 1,050,000.
- Which ranks higher, Claude Opus 5 or GPT-5.6 Luna?
- Claude Opus 5 ranks higher on the public arena (Claude Opus 5 #8, GPT-5.6 Luna #36). Ranks move as the leaderboard updates.
- Can I use Claude Opus 5 and GPT-5.6 Luna with the same API key?
- Yes. On FastMetal's OpenAI-compatible endpoint, switch by passing "anthropic-claude-opus-5" or "gpt-5.6-luna" as the model. Each is billed at its own rate from the same prepaid balance.
Try both on one API key
Create an account and add credit to call Claude Opus 5 and GPT-5.6 Luna from the browser chat and the API. No monthly fee.