Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.ruoli.dev/llms.txt

Use this file to discover all available pages before exploring further.

Token quota and account balance are separate. Each token has its own usage limit — even if your account balance is sufficient, you’ll get an error when the token quota is exhausted.Solution: Edit the token in the console to increase the quota or enable “unlimited quota”.
The model you requested is not in the current token’s group, or all channels in that group are unavailable.Solution: Create a new token with a group that includes the model, or try a different model.
  • To save money: cn (0.01)
  • For large context: gemini-ant (0.2, 1M), codex-team (0.2, 1M), or claude-ant (0.6, 1M)
  • For official quality: claude-officially (1.6) or gemini-officially (1)
See Models & Groups for details.
  • Check that the API Key is correct (watch for leading/trailing spaces)
  • Confirm the Key hasn’t been deleted or disabled
  • Confirm the request header format: Authorization: Bearer sk-YOUR-KEY
The multiplier is a pricing coefficient — the lower the multiplier, the cheaper it is. For example, a multiplier of 0.1 means you pay only one-tenth of the official price. See Models & Groups for details.
RuoLi is compatible with the OpenAI API format and supports all OpenAI-compatible tools, including:
To use third-party models like gpt-5.5, GLM, or Kimi, open CC Switch → Edit Provider → Model Mapping and point each Claude alias at the target model — no JSON editing needed.See Switch Models in Claude Code for details.
It’s almost always the default attribution header — every request carries a dynamic session ID that blows out the upstream prompt cache.In CC Switch → Edit Provider → Config JSON, add one line to env:
"CLAUDE_CODE_ATTRIBUTION_HEADER": "0"
Recommended whether you’re on native Claude or a third-party model. See Disable Attribution Header for details.
This usually means the conversation context has exceeded the model’s context window limit.Solutions:
  • Use the /compact command in Claude Code to manually compress context
  • Configure auto-compression: add "CLAUDE_AUTOCOMPACT_PCT_OVERRIDE": "50" to the env block in ~/.claude/settings.json — lower values trigger compression earlier
  • Choose a group with larger context support, such as claude-ant (1M context) or codex-team (1M context)
See Claude Code context compression for details.
Unified URL: https://ruoli.dev/v1The exact format varies slightly by tool — see each tool’s configuration page for details.