Rate Limit Errors¶
What is a Rate Limit?¶
RagLeap Manager AI uses ~7,500 tokens per message. Some AI model tiers limit tokens per minute (TPM). When this limit is hit, Manager AI shows a rate limit warning.
Which Models Have This Problem?¶
| Model | TPM Limit | Works with RagLeap? |
|---|---|---|
| GPT-4.1 standard | 30K/min | Hits limit after 3-4 messages |
| GPT-4o-mini | 200K/min | Recommended for OpenAI users |
| gemini-2.5-flash | 1M/min | Best — free, no limits |
| gemini-1.5-flash | 1M/min | Great free option |
| claude-haiku-4-5-20251001 | High | Fast and efficient |
Fix — Switch Your Model¶
Tell Manager AI: Switch my model to gemini-2.5-flash
Or go to Settings → AI Settings → change model → Save
Still Hitting Limits?¶
- Upgrade your API plan with your provider
- Wait 60 seconds — TPM resets every minute
- Switch to a different provider