Standard Compute

LLM API Troubleshooting

Fix the API errors that break your agent

Clear, accurate fixes for the LLM API errors developers hit most — 429 rate limits, “quota exceeded”, context length, overloaded servers, and the agent-specific ones that stall OpenClaw and Hermes. Causes, step-by-step fixes, and how to stop hitting them.

Rate limits

OpenAI “Rate limit reached for requests” (429)

OpenAI tokens-per-minute (TPM) rate limit

429Any provider

429 Too Many Requests — what it means & how to fix

429Anthropic (Claude)

Anthropic / Claude rate limit exceeded (429)

529Anthropic (Claude)

Anthropic “Overloaded” error (529)

OpenAI “The server is overloaded” (503)

OpenClaw “API rate limit reached” (429)

Why your AI agent keeps getting rate limited (and how to stop it)

OpenRouter 429 “rate limited” — causes, fixes and the structural way out

Gemini CLI “Quota exceeded” (429) — how to fix

Cline “API Request Failed” 429 — how to fix

429Hermes Agent

Hermes Agent keeps hitting rate limits — how to fix

429OpenCode / upstream

OpenCode rate limited / usage limits — which provider is stopping you & the flat-rate fix (2026)

Quota & billing

OpenAI “You exceeded your current quota” (insufficient_quota)

OpenAI monthly usage / billing hard limit reached

“Quota exceeded — please use your own API key” explained

Cursor “You've hit your usage limit” — how to fix

Claude Code usage limit reached — weekly limits, fixes & the alternatives heavy users switch to (2026)

AI agent burning through API credits — why it happens and the permanent fix

Cline API costs too high — how to cut the bill without losing frontier models

Codex “You've hit your usage limit. Upgrade your plan or add credits” — the Fixes (2026)

GitHub Copilot premium requests exhausted — how to fix

Windsurf out of credits — how to fix

Claude Weekly Limit — Why It Reset Early, When It Resets, and How to Keep Working

OpenCode / Anthropic

Can you use a Claude Pro/Max subscription with OpenCode? Login, limits & alternatives (2026)

Roo Code API costs too high — reduce spend or go flat rate (2026)

Pi coding agent API costs — already lean, here's how to make them fixed (2026)

Model limits

400Any provider

“Maximum context length exceeded” — what it means & how to fix

404OpenAI / OpenAI-compatible

“The model does not exist or you do not have access”

Auth

401OpenAI / OpenAI-compatible

“Incorrect API key provided” (401) — how to fix

Agent errors

OpenClaw “conflicting plugin install metadata” (shared SQLite state) — how to fix

Tired of fixing rate limits?

Most of these errors share one root cause: per-token, per-minute metering. Standard Compute is flat-rate, no per-minute caps, and automatic failover — so agents degrade gracefully instead of throwing 429s. See what you'd pay → · How it connects →