How to Use GPT-4o in Claude Code
GPT-4o handles code, multilingual text, and structured extraction. OpenAI currently maps this undated name to GPT-4o 2024-08-06; use a dated name to identify the revision you tested.
Prompts to try with GPT-4o in Claude Code
Copy an example into Claude Code, or ask in your own words.
Use GPT-4o to review the current diff for incorrect behavior. Read the relevant surrounding code, give a concrete failure example for each finding, and propose a focused test.
Use GPT-4o to translate the project's English interface strings into Spanish. Follow the existing glossary, preserve interpolation tokens, and list ambiguous source strings separately.
Ask GPT-4o through Spiral to review the diff on this branch and list the three riskiest changes. Save the result in ./assets.
Get a second opinion from GPT-4o on this function, then tell me what the call cost.
Summarize these release notes with GPT-4o in five bullet points.
How to set up GPT-4o in Claude Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
This adds Spiral to the current project. Add --scope user to use it in every project.
claude mcp add --transport http spiral https://api.spiral.ai/mcp
- Sign in
Start claude, type /mcp, choose spiral and Authenticate. Sign in to Spiral in the browser and click Authorize. No API key.
Use the Spiral skill
- Send Claude Code this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run GPT-4o
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openai", "model": "openai/gpt-4o", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Claude Code finds GPT-4o on Spiral, shows the price and runs it.
GPT-4o providers and prices
Spiral runs GPT-4o through OpenAI. Open a card for today's price, with every fee included.
Up to $0.053363 per run at the default inputs. $3.25 per 1M input tokens, $1.625 per 1M cached input tokens, $13.00 per 1M output tokens.
Settings you can ask for
Ask Claude Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top p | 0 to 1 | ||
| Logprobs | on or off | ||
| Max tokens | a whole number from 16 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Parallel tool calls | on or off |
Every API field of GPT-4o is on its tool page: open its card above.
What can you do with GPT-4o in Claude Code?
GPT-4o can turn project code and written requirements into review findings, patches, and test drafts. Give it the relevant files, expected behavior, and a precise definition of the result.
Extract action items from meeting notes
Turn notes into records containing the task, owner, due date, and supporting passage. Mark missing owners or dates as unknown.
Review a proposed code change
Ask GPT-4o to examine a focused diff, explain likely failures, and propose tests for each finding. Give it the relevant surrounding code and expected behavior.
Draft multilingual customer replies
Provide the customer message and approved response points, then request a reply in the customer's language. Specify tone and preserve product names.
Classify incoming requests
Define categories such as bug report, feature request, and how-to question. Supply labeled examples and require a supporting passage for each classification.
Convert a brief into a structured draft
Turn an editorial brief into a draft with a specified audience, outline, and length. Use GPT-4o 2024-11-20 when its documented creative writing improvement matters to the task.
Translate product instructions
Translate approved instructions with a supplied glossary. Ask for a separate list of ambiguous phrases so a reviewer can resolve them before publication.
Turn requirements into test cases
Provide expected behavior and examples, then request normal, boundary, and failure cases. Require each proposed test to identify the requirement it checks.
More on GPT-4o and Claude Code
Overview
GPT-4o in other agents
More Claude Code guides
Questions about GPT-4o in Claude Code
Can Claude Code use GPT-4o?
Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run GPT-4o like any built-in tool.
How much does GPT-4o cost in Claude Code?
Up to $0.053363 per run at the default inputs. $3.25 per 1M input tokens, $1.625 per 1M cached input tokens, $13.00 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Claude Code plan is separate.
Do I need an OpenAI account to use GPT-4o in Claude Code?
No. Spiral runs GPT-4o on its own OpenAI account. You sign in to Spiral once and never paste a key.
How long does it take to set up GPT-4o in Claude Code?
About two minutes: add the hosted server, then sign in.
Does Claude Code wait while GPT-4o runs?
GPT-4o runs in the background. Claude Code starts the run, checks on it and gives you the text, json when it's ready.
How much can GPT-4o read and write?
GPT-4o has a 128,000-token context window and can generate up to 16,384 tokens in one response. Treat those as separate limits when planning long-document work.
Does schema-matched JSON mean the answer is correct?
No. Structured Outputs controls the response structure, but the values and conclusions can still be wrong. Check extracted details against the source.
Can GPT-4o work with languages other than English?
Yes. OpenAI reported improved non-English text performance compared with GPT-4 Turbo. For localization, specify the target language, audience, and terms that must stay unchanged.
What should a useful GPT-4o prompt include?
State the task, relevant source material, required format, and what to do when information is missing. Include a few input/output examples when the expected pattern is hard to describe.
How do the GPT-4o names differ?
OpenAI currently maps GPT-4o to GPT-4o 2024-08-06. GPT-4o 2024-08-06 introduced Structured Outputs, while GPT-4o 2024-11-20 adds a documented creative writing improvement.
Start using GPT-4o in Claude Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Claude Code:
set up https://spiral.ai/SKILL.md


