feat: usage of new keep_alive parameter in ollama #275

New Issue

GiteaMirror · 2025-11-11T14:15:00-06:00

GiteaMirror commented

2025-11-11 14:15:00 -06:00

Originally created by @nick-tonjum on GitHub (Feb 7, 2024).

Hi! Ollama recently updated and added the following:

keep_alive parameter: control how long models stay loaded
When making API requests, the new keep_alive parameter can be used to control how long a model stays loaded in memory:

curl http://localhost:11434/api/generate -d '{
  "model": "mistral",
  "prompt": "Why is the sky blue?",
  "keep_alive": "30s"
}'
If set to a positive duration (e.g. 20m, 1hr or 30), the model will stay loaded for the provided duration
If set to a negative duration (e.g. -1), the model will stay loaded indefinitely
If set to 0, the model will be unloaded immediately once finished
If not set, the model will stay loaded for 5 minutes by default

I believe this would be great to have in the 'Advanced' tab in ollama-webui's settings, for someone who regularly uses the same model I hate having to wait for it to load after it unloads after sitting idle. I can also make a pull request to implement this when I have free time too, I'm just very busy for this next week.

Thanks!

Originally created by @nick-tonjum on GitHub (Feb 7, 2024). Hi! Ollama recently updated and added the following: ``` keep_alive parameter: control how long models stay loaded When making API requests, the new keep_alive parameter can be used to control how long a model stays loaded in memory: curl http://localhost:11434/api/generate -d '{ "model": "mistral", "prompt": "Why is the sky blue?", "keep_alive": "30s" }' If set to a positive duration (e.g. 20m, 1hr or 30), the model will stay loaded for the provided duration If set to a negative duration (e.g. -1), the model will stay loaded indefinitely If set to 0, the model will be unloaded immediately once finished If not set, the model will stay loaded for 5 minutes by default ``` I believe this would be great to have in the 'Advanced' tab in ollama-webui's settings, for someone who regularly uses the same model I hate having to wait for it to load after it unloads after sitting idle. I can also make a pull request to implement this when I have free time too, I'm just very busy for this next week. Thanks!

GiteaMirror closed this issue

2025-11-11 14:15:01 -06:00

GiteaMirror commented

2025-11-11 14:15:02 -06:00

@justinh-rahb commented on GitHub (Feb 7, 2024):

Duplicate: #596

@justinh-rahb commented on GitHub (Feb 7, 2024): Duplicate: #596

GiteaMirror referenced this issue

2025-11-11 17:15:04 -06:00

[PR #275] [MERGED] Docker compose modularity improvements + kubernetes support #7034

Sign in to join this conversation.