FAQ – AskAI Failures Caused by Token Limits, Context Size, or Model Overload (only on Odoo SH and On Premise)

(General guidance. Please note that if you are experiencing a similar behaviour in Odoo SaaS, the behaviour is not normal and you should report it to the support team.)

As a reminder, Odoo AI is only available in the Enterprise version (paid), not in Community.

1. What are the common symptoms?

Users may report one or more of these messages:

• “Sorry, I couldn’t process your request right now. Please try again later.”
  
• Invalid Operation – “The model is overloaded. Please try again later.”
  
• Errors appear regardless of model (GPT / Gemini, Lite or Pro).
  
• Simple queries (CRM, invoices, leads, etc.) intermittently fail.
  

These symptoms are typical when the AI provider rejects the request due to token/context limits or throughput restrictions.

2. Why does AskAI fail even with simple questions?

Even simple questions can generate large internal prompts when the agent has:

• Assigned topics
  
• Multiple tools enabled
  
• Database-heavy queries
  

AskAI sends a combined payload containing:

• User prompt
  
• Agent system prompt
  
• Tool prompts
  
• Extracted database data
  
• Model-specific tokenization
  

This can result in very high token counts, exceeding the customer’s API tier.

3. Why does it work sometimes but not always?

Because the token usage varies per request.

If the request is near the API token limit, some calls stay below the cap and succeed, while others exceed it and fail.

This produces intermittent failures, identical to what Synkiria observed.

4. How do we confirm the issue?

Check the Odoo server logs.

Every AI call logs the token count.

Example from a real case (Synkiria):

• Tokens sent: 39,233
  
• API limit: 30,000 (OpenAI Tier 1)
  → The provider rejects the request → AskAI returns a generic error.
  

Any log showing tokens above the provider’s limit confirms the cause.

5. Why does Odoo show a generic “model overloaded” message?

If the AI provider rejects the request before processing it, Odoo receives no detailed error code.

The fallback message (“model is overloaded” / “cannot process now”) is shown automatically.

This is normal when hitting token caps, rate limits, or context overflows.

6. What can be done to fix it?

✔ Upgrade the API tier

As an example, for OpenAI:

GPT 4o Tier 1 → 30k tokens/s (often insufficient for multi-topic AskAI agents). You can always check directly with your AI provider for their tier plans since they change according to model and you might see different pricing depending on when you are reading this post.

✔ Use a more efficient model

Recommended: Gemini Flash 2.5

• Lower token usage
• Faster
• Higher quality
• More cost-efficient
  Requires a Google Gemini API key.

✔ Reduce complexity of agent topics (optional mitigation)

Fewer topics → fewer tools → fewer tokens.

7. Quick Support checklist

Odoo Enterprise users can create a support ticket or contact support directly. Before contacting support, please confirm:

✔ Agent has topics

✔ You are on a low API tier

✔ Recommend Gemini Flash 2.5 or upgrading the API tier

If all conditions match → This is a token-limit case.