open-webui / open-webui/computer
bug: Thinking blocks broken with Deepseek
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 569
- Forks
- 79
- PR merge metrics
- No merged PRs in 30d
Description
When using Deepseek V4 (via Opencode Go) thinking blocks are split every 1-3 words, resulting in hundreds of tiny blocks per turn. Other models seem to work fine (eg. GLM).
cptr 0.9.21
Debian Trixie
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the issue with Deepseek V4 through Opencode Go on Debian Trixie, then compare the resulting thinking blocks with GLM. Trace the thinking-block handling and identify why Deepseek content is split every few words. Done means a Deepseek turn produces coherent thinking blocks rather than hundreds of tiny blocks, while other models continue to work.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai, backend
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100