- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello everyone,
From keeping up with development in Optimum-Intel and OpenVINO GenAI I know that gpt-oss architecture is supported in OpenVINO and that performance is expected to improve by 2025.4. I am SearchSavior.
However, obtaining OpenVINO IR and inferencing returns only "!!!!!!!" until max_new_tokens. See this issue from mwrothbe. We have been tinkering with gpt-oss-20b on and off for a few weeks, especially after this PR was merged into OpenVINO GenAI without progress. Yet while development continues we have no idea where to follow.
Maybe the documentation should be updated if gpt-oss support is still baking?
Thank you,
Emerson
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Echo9Zulu,
Thank you for reaching out.
Since this is the same issue as the one on GitHub #2984, we will continue the investigation there and close this thread. Are you okay with this arrangement?
Regards,
Zul
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Zul,
Sure, that sounds good. Look forward to collaborating on this issue.
Emerson.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thank you for your question. If you need any additional information from Intel, please submit a new question, as this thread is no longer being monitored.
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page