feat: verbose output (show timings per response) #74

New Issue

2025-11-11T14:03:58-06:00

GiteaMirror commented

2025-11-11 14:03:58 -06:00

Originally created by @johncadengo on GitHub (Dec 7, 2023).

When running Ollama from the command prompt, you can type the --verbose argument to get timings that output like this:

$ ollama run --verbose llama2
>>> Hi
Hello! It's nice to meet you. Is there something I can help you with or would you like to chat?

total duration:       1.279118717s
load duration:        891.933µs
prompt eval count:    21 token(s)
prompt eval duration: 231.416ms
prompt eval rate:     90.75 tokens/s
eval count:           25 token(s)
eval duration:        1.042407s
eval rate:            23.98 tokens/s

Is there a way to get this to appear in the web UI?

Originally created by @johncadengo on GitHub (Dec 7, 2023). When running Ollama from the command prompt, you can type the `--verbose` argument to get timings that output like this: ``` $ ollama run --verbose llama2 >>> Hi Hello! It's nice to meet you. Is there something I can help you with or would you like to chat? total duration: 1.279118717s load duration: 891.933µs prompt eval count: 21 token(s) prompt eval duration: 231.416ms prompt eval rate: 90.75 tokens/s eval count: 25 token(s) eval duration: 1.042407s eval rate: 23.98 tokens/s ``` Is there a way to get this to appear in the web UI?

GiteaMirror closed this issue

2025-11-11 14:03:58 -06:00

GiteaMirror commented

2025-11-11 14:03:59 -06:00

@tjbck commented on GitHub (Dec 7, 2023):

Hi, I'll think of a way to incorporate this into the webui, stay tuned!

@tjbck commented on GitHub (Dec 7, 2023): Hi, I'll think of a way to incorporate this into the webui, stay tuned!

GiteaMirror commented

2025-11-11 14:03:59 -06:00

@tjbck commented on GitHub (Dec 19, 2023):

The requested feature has been added with #215. Please try out the latest version, and let me know if you encounter any issues. Thanks!

@tjbck commented on GitHub (Dec 19, 2023): ![image](https://github.com/ollama-webui/ollama-webui/assets/25473318/49b833ed-40da-43a2-ba85-0e6666860e3e) The requested feature has been added with #215. Please try out the latest version, and let me know if you encounter any issues. Thanks!

GiteaMirror commented

2025-11-11 14:03:59 -06:00

@johncadengo commented on GitHub (Dec 21, 2023):

Ty! 👏 👏 👏

@johncadengo commented on GitHub (Dec 21, 2023): Ty! 👏 👏 👏

GiteaMirror commented

2025-11-11 14:03:59 -06:00

@johncadengo commented on GitHub (Apr 22, 2024):

@tjbck I have just spun up the latest version of this image and so many things have changed. Do you happen to know if this feature is still present? I can't seem to find it in the latest version.

@johncadengo commented on GitHub (Apr 22, 2024): @tjbck I have just spun up the latest version of this image and so many things have changed. Do you happen to know if this feature is still present? I can't seem to find it in the latest version.

GiteaMirror commented

2025-11-11 14:04:00 -06:00

@tjbck commented on GitHub (Apr 22, 2024):

All still there :)

@tjbck commented on GitHub (Apr 22, 2024): <img width="886" alt="image" src="https://github.com/open-webui/open-webui/assets/25473318/c95022ff-141d-4435-9637-c09bbd366657"> All still there :)

GiteaMirror commented

2025-11-11 14:04:01 -06:00

@johncadengo commented on GitHub (Apr 24, 2024):

@tjbck okay, thanks for showing that.

I think what happened between versions for me is that I was decoupling the model API from the docker container and trying to have a separate API for the models. At first, I was using the LiteLLM models feature. I'm assuming it doesn't work for that?

Anyhow, I figured out how to use Ollama models instead and now I see the generation info again.

@johncadengo commented on GitHub (Apr 24, 2024): @tjbck okay, thanks for showing that. I think what happened between versions for me is that I was decoupling the model API from the docker container and trying to have a separate API for the models. At first, I was using the LiteLLM models feature. I'm assuming it doesn't work for that? Anyhow, I figured out how to use Ollama models instead and now I see the generation info again.