cloudflare / cloudflare/ai

AI Gateway REST: google/gemini-3.1-flash-lite failing all requests with HTTP 500

Open
#616 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
TypeScript
Stars
1.2k
Forks
345
Avg merge
13h 31m
Merged PRs (30d)
1

Description

All AI Gateway REST `/ai/run` requests for `google/gemini-3.1-flash-lite` have been failing with HTTP 500 for over 7 days (Since July 6th).

Reproduction:
Use the provided example CURL on [Cloudflare's official model page](https://developers.cloudflare.com/ai/models/google/gemini-3.1-flash-lite/#API-Schemas):
```
curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
--header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
--header "Content-Type: application/json" \
--data '{
"model": "google/gemini-3.1-flash-lite",
"input": {
"contents": [
{
"parts": [
{
"text": "What are the three laws of thermodynamics?"
}
],
"role": "user"
}
]
}
}'
```

HTTP 500 Response:
```
{
"errors": [
{
"message": "Model execution failed (Error)",
"code": 2002
}
],
"success": false,
"result": {},
"messages": []
}
```

Cloudflare support has failed to raise an outage for this issue, or respond to tickets regarding it.

Contributor guide

Open the contributing guide

Research direction

Start by reproducing the failure through the AI Gateway REST `/ai/run` entry point with the Cloudflare model-page cURL example. Trace the request handling for `google/gemini-3.1-flash-lite` and compare it with a working model; done means these requests return a successful model response instead of HTTP 500 with error code 2002.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
ai, api
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.