setup re-run with --vision gpu does not add --vision to an existing config's args
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 76/100
Research direction
Start in setup.py and compare the fresh-config path with the rerun path for --vision gpu. Check how existing args are preserved and how the vision section is written; reproduce by rerunning the setup command from the issue against an existing config. Done means the config includes the vision engine flags and image requests no longer fail with the reported startup error.
Written by the indexing model from the issue text.
Description
What happened
On a machine whose strata-*.json was created earlier without images, re-running setup to enable vision:
./setup.sh --setup --model IQ3_S --vision gpu --context 262144 --yes
writes the vision section into the config but leaves args without --vision --vram-reserve-mib 700. The server then starts the engine without vision and every image request fails with:
this engine was started without --vision
while /v1/models still advertises input_modalities: ["text", "image"] (taken from the vision section), which is misleading.
Workaround
Manually append --vision --vram-reserve-mib 700 to the config's args and restart. Verified working on two nodes (RTX 5090, RTX PRO 6000).
Suspected cause
In setup.py, args += ["--vision", "--vram-reserve-mib", ...] only runs on the fresh-config path. The re-run path preserves the existing args and only adds the vision section, so the engine never gets the flag.
Environment
- Strata at commit
82f46a8(engine 0.1.38 / 0.1.40, local build, CUDA 13.2, sm_120) - Ubuntu 24.04, RTX 5090 32GB / RTX PRO 6000 96GB
- Qwen3.8-Flash-Next IQ3_S
- Dominant language
- C++
- Stars
- 11.6k
- Forks
- 1k
- Avg merge
- 7h 46m
- Merged PRs (30d)
- 30
Getting set up
This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.
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.
More from Niko1221/Strata
-
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Maintainers usually reply within 1 day
-
expert_cache_segmented_test fails on HIP builds instead of skipping (--vram-elastic is CUDA-only)Possibly taken @nekomario28 claimed this today. Open
Difficulty 2/5 1-3 hours Newbie friendliness 83/100
Maintainers usually reply within 1 day
-
hip_q2_zero fails on gfx1201 (R9700) with ROCm 7.10: Q2_0 signed-zero fix e9a5f8d is gated to gfx1012 / HIP < 7Possibly taken @nekomario28 claimed this today. Open
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 72/100
Niko1221/Strata#1463 · 2 comments ·
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 82/100
Maintainers usually reply within 1 day
Similar issues
-
Status: Awaiting triage
Difficulty 2/5 1-3 hours Newbie friendliness 75/100
espressif/arduino-esp32#12984 ·
Maintainers usually reply within 1 day
-
torch_ops/logprob.cu does not compile with the serving container's nvcc (13.3.73); check_torch_ops.py cannot run as shippedPossibly taken A pull request linked to this issue is open or already merged. Open
Difficulty 2/5 Under an hour Newbie friendliness 72/100
ashhart/TensorFold#535 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 1 day
-
agent:Windows bug
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 1 day