Glossary
Base URL: where your client connects
A base URL is the address root a client uses to reach an API. Changing two values in a client — base URL and key — is usually all a migration takes.
Updated
Two roots, two protocols
OpenAI-shaped clients point to https://llmcode.io/v1 and call /chat/completions, /responses, /embeddings and /models.
Anthropic-shaped clients point to https://llmcode.io without /v1; the SDK appends /v1/messages itself and sends the anthropic-version header.
Why the trailing /v1 matters
The /v1 segment is part of the URL for OpenAI-shaped clients. Omitting it or doubling it produces a 404.
The Anthropic SDK expects the base without /v1 and adds the version segment itself.
One key across both
The same key works on both roots and across models. You change the base URL once per client, not per model.
Detailed snippets for cURL, Python and TypeScript are on /instructions and on each model page.
Published model prices
USD per 1M tokens. The numbers come from the public catalog on every page rebuild — this is the sale price, not an estimate.
| Model | Input | Output | Cache |
|---|---|---|---|
| Claude Opus 5.5 | $2.70 | $8.00 | $0.14 |
| DeepSeek Flash | $0.50 | $2.00 | $0.01 |
| Claude Sonnet 5.5 | $1.33 | $6.67 | $0.07 |
| DeepSeek V4 Pro | $2.20 | $6.60 | $0.07 |
Top-up from $10, balance in USD. 1 USD = 1,000,000 UT. No subscription and no auto-charge.
Questions
Is the base URL https or http?
HTTPS only. http://llmcode.io redirects to https.
Do I add /v1 for Anthropic clients?
No. Point the Anthropic SDK at https://llmcode.io and it appends /v1/messages.
Why does the model page show the same URL?
Because the endpoint is shared: the model is chosen in the request body, not in the URL.
Can I self-host the gateway?
No. The gateway is operated by LLMcode; you connect to it over the public URL.