Erratic behavior after several prompts
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 1.9k
- Forks
- 152
- Avg merge
- 4h 14m
- Merged PRs (30d)
- 11
Description
Finding consistently that after 10 or so prompts qwen3.5-4b-flm starts returning empty responses, and then intermittently goes to 99% npu indefinitely (> 5 minutes when previously similar prompts had taken 20-25 seconds) and need to end task on flm.exe to regain control of npu.
running under lemonade server and lemonade sometimes "loses control" of flm when this happens - I have raised a separate issue at lemonade's github where I mentioned this.
Version
10.0.1
OS
Microsoft Windows 11 Pro 10.0.26200 (Build 26200)
CPU
AMD RYZEN AI MAX+ 395 w/ Radeon 8060S (16 cores, 32 logical processors)
GPU
AMD Radeon(TM) 8060S Graphics
Contributor guide
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
Start by reproducing qwen3.5-4b-flm under Lemonade server on Windows 11 with the listed Ryzen AI Max+ 395 and Radeon 8060S hardware. Run roughly ten prompts and observe the empty responses, sustained 99% NPU usage, and recovery after ending flm.exe. Done should mean repeated prompting no longer causes the response failure or indefinite NPU utilization.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- ai-infra-agents
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 42/100