genai timeout ollama

[divemasterjm]

Describe the problem you are having

genai using ollama takes 2-3 minutes (i don't have gpu), is possible to define a time out? i use it in another programs and it works with the timeout.

Version

0.15

What browser(s) are you using?

No response

Frigate config file

genai:
  enabled: true
  provider: ollama
  base_url: http://192.168.8.38:11434
  model: llava

Relevant Frigate log output

no log error

Relevant go2rtc log output

no errors

FFprobe output from your camera

it not related to a camera

Frigate stats

No response

Install method

Docker Compose

docker-compose file or Docker CLI command

not relevant

Object Detector

Coral

Network connection

Wired

Camera make and model

dahua

Screenshots of the Frigate UI's System metrics pages

No response

Any other information that may be helpful

No response

[leccelecce]

I had a similar issue and a response from the maintainers that may help you is here: #14472 (reply in thread)

[divemasterjm]

thanks!.

[hawkeye217]

The default timeout is 60 seconds.

frigate/frigate/genai/__init__.py

Line 28 in 2d27e72

def __init__(self, genai_config: GenAIConfig, timeout: int = 60) -> None:

CPU inference with Ollama is not recommended. Inference times longer than a couple of seconds make genai completely impractical.

[divemasterjm]

understood, thanks.