Guide
Claude API pricing through an Anthropic-compatible endpoint
LLMcode accepts Anthropic Messages requests and serves Claude models from its catalog. You pay the published per-token price; the exact numbers are in the table below.
Updated
Connecting an Anthropic client
Set the base URL to https://llmcode.io without /v1 — the SDK appends /v1/messages. Authenticate with x-api-key and send anthropic-version: 2023-06-01.
The same key also works for OpenAI-compatible endpoints. Pick the model per request.
Price, cache and limits
Claude profiles have input, output and cached-input prices. Caching applies only when the provider reports cached tokens; the gateway cannot force it.
A key can carry a spend limit in UT, an IP allowlist and an expiry. Requests above the balance return 402.
Honest limits
This is compatible access through a partner channel, not a direct Anthropic account. Latency and upstream limits can differ from the native service.
Streaming works for messages. Responses streaming is disabled in this version.
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
Which base URL does the Anthropic SDK need?
https://llmcode.io without /v1. The request goes to POST /v1/messages.
Does Claude Code work?
Yes: set ANTHROPIC_BASE_URL, ANTHROPIC_AUTH_TOKEN and the ANTHROPIC_DEFAULT_*_MODEL variables. Check with claude /status.
Is the price the same as Anthropic's?
No. The sale price is LLMcode's published profile price per 1M tokens, shown in the catalog.
Can I use one key for Claude and DeepSeek?
Yes. A key is not bound to a model; the model is chosen in each request.