Is Prompt Playground Available for Self-Hosted Phoenix?

Mr T. · 2025-12-03 07:48:45.002756+00

or is the prompt playground not available for the self-hosted Open Source Version of phoenix?

Is Prompt Playground Available for Self-Hosted Phoenix? | Arize AI Community

11 comments

· Sorted by Oldest

RunLLM
APP
·
The prompt playground feature is available in the Phoenix platform, but the specific availability for the self-hosted open-source version is not explicitly mentioned in the provided data sources. The feature was introduced in a GitHub pull request related to the playground feature, which suggests it is part of the platform's capabilities. However, the documentation does not clearly state whether this feature is included in the self-hosted version.Would you like more details on how to set up or use the prompt playground in Phoenix?Sources: - GitHub pull request
💡 Hint: Mention RunLLM in the thread for followups.
Dustin N.
·
Hi Mr T. Thanks for trying out Phoenix! We’re actively working on the prompt playground feature for Phoenix and hope it will be released soon
Mr T.
·
Dustin N. thanks for answering so quickly. If it is not released yet, is the feature branch available where its being developed so i could try it out at its current state?
Dustin N.
·
It’s not fully ready yet, we’ll make an announcement when it is!
Mr T.
·
I saw lots of playground related commits in the main branch, which is why I thought the feature is already implemented. Is there a experimental flag that I could turn on to access it? I understand its in active development... The reason why I'd like to get access to it anyway even if its still buggy is that our team is now evaluating the Logging Framework that we will work with for the next few years. And the playground would be a big plus point for phoenix if I could include it in our evaluation
Dustin N.
·
I see, Anthony P. would you say the front end is ready for an initial evaluation in its current state?
Dustin N.
·
Give us a little bit and we can send you details over DM depending on where we’re at
Mr T.
·
I fully understand it to be WIP... even if functionality is missing, just being able to roughly gauge how it can be used in the future would be enough for now
Anthony P.
·
For surface level evaluation using openai I don't see why not
👍1
Anthony P.
·
I will DM you shortly Mr T.
Mr T.
·
Anthony P. Thank you so much, really looking forward to it.

RunLLM
APP
·
The prompt playground feature is available in the Phoenix platform, but the specific availability for the self-hosted open-source version is not explicitly mentioned in the provided data sources. The feature was introduced in a GitHub pull request related to the playground feature, which suggests it is part of the platform's capabilities. However, the documentation does not clearly state whether this feature is included in the self-hosted version.Would you like more details on how to set up or use the prompt playground in Phoenix?Sources: - GitHub pull request
💡 Hint: Mention RunLLM in the thread for followups.
Dustin N.
·
Hi Mr T. Thanks for trying out Phoenix! We’re actively working on the prompt playground feature for Phoenix and hope it will be released soon
Mr T.
·
Dustin N. thanks for answering so quickly. If it is not released yet, is the feature branch available where its being developed so i could try it out at its current state?
Dustin N.
·
It’s not fully ready yet, we’ll make an announcement when it is!
Mr T.
·
I saw lots of playground related commits in the main branch, which is why I thought the feature is already implemented. Is there a experimental flag that I could turn on to access it? I understand its in active development... The reason why I'd like to get access to it anyway even if its still buggy is that our team is now evaluating the Logging Framework that we will work with for the next few years. And the playground would be a big plus point for phoenix if I could include it in our evaluation
Dustin N.
·
I see, Anthony P. would you say the front end is ready for an initial evaluation in its current state?
Dustin N.
·
Give us a little bit and we can send you details over DM depending on where we’re at
Mr T.
·
I fully understand it to be WIP... even if functionality is missing, just being able to roughly gauge how it can be used in the future would be enough for now
Anthony P.
·
For surface level evaluation using openai I don't see why not
👍1
Anthony P.
·
I will DM you shortly Mr T.
Mr T.
·
Anthony P. Thank you so much, really looking forward to it.