📱 Client Type

Other

💻 Operating System

Other

📦 Deployment Platform

Other

📌 Version

latest (static analysis — not version-specific)

🌐 Browser

Chrome, Other

🐛 What happened?

executeWithRetry in packages/model-runtime/src/providers/githubCopilot/index.ts reads retry-after correctly but caps it at 10 seconds:

Math.min(retryAfter, 10_000)

A retry-after of 3 hours (quota exhaustion) gets truncated to 10 seconds and retried 3 times instead of surfacing
immediately as unrecoverable.

📷 How to reproduce it?

Static analysis finding in src/providers/githubCopilot/index.ts

Trigger a quota exhaustion 429 with retry-after > 300s.

Observe that executeWithRetry caps the delay at 10s and retries MAX_RATE_LIMIT_RETRIES times instead of surfacing.

🚦 What it should be?

retry-after above a threshold should surface immediately as quota exhaustion, not consume retry attempts.

Suggested fix:
const STOP_THRESHOLD_MS = 5 * 60 * 1000
if (retryAfter && retryAfter > STOP_THRESHOLD_MS) {
throw this.mapError(error)
}

📝 Additional Information

Documented across production systems:
github.com/SirBrenton/pitstop-truth

🛠️ Willing to Submit a PR?

Yes, I am willing to submit a PR

✅ Validations

• Read the docs.
• Check that there isn't already an issue that reports the same bug to avoid creating a duplicate.
• Make sure this is a LobeChat issue and not a third-party library or provider issue.
• Check that this is a concrete bug. For Q&A, please use GitHub Discussions or join our Discord Server.