[Other] tools/test_strata_mcp.py: the Mac DeepSeek list test fails on main (it still expects the two IST-DAS sets)
Maintainers usually reply within 3 days
@XiaoBinGan is already working on this.
Since Nov 1, 2026.
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 78/100
Research direction
Start with the failing test test_deepseek_on_a_mac in tools/test_strata_mcp.py around line 787, then check the Mac set contents in strata_control/dsv41_fit.py (MAC_SET_PACK, MAC_MEASURED_PACK, MAC_NOT_YET). Run python -m pytest tools/test_strata_mcp.py -q -k test_deepseek_on_a_mac to reproduce. Done when the expected list names the four offered sets and the whole file passes on main.
Written by the indexing model from the issue text.
Description
tools/test_strata_mcp.py is red on main (c62d5e3a) because the Mac DeepSeek check still expects the two
IST-DAS sets.
Reproduce
$ python -m pytest tools/test_strata_mcp.py -q -k test_deepseek_on_a_mac
AssertionError: Lists differ: ['IST-DAS-2.00bpw', 'IST-DAS-2.25bpw', 'UD-IQ1_S', 'UD-IQ1_M']
!= ['IST-DAS-2.00bpw', 'IST-DAS-2.25bpw']
tools/test_strata_mcp.py:787
1 failed, 52 deselected
Python 3.11.16, Windows 11; the test mocks an Apple Silicon Mac, so the host OS does not matter. It is the only
failure in the file on unmodified main (python -m pytest tools/test_strata_mcp.py -q: 1 failed, 51 passed).
Why
The expectation predates the Unsloth 1-bit DeepSeek sets being offered on a Mac:
strata_control/dsv41_fit.py MAC_SET_PACK now holds UD-IQ1_S and UD-IQ1_M as well, and MAC_NOT_YET is empty
by design ("the Mac gate stays as an empty mechanism", commit 33bbbca4; UD-IQ1_S came with its M5 Max line in
6a7d7505). The model list therefore offers four sets, and their Mac words are all present, including UD-IQ1_M's
"Not run on a Mac yet" line from MAC_MEASURED_PACK.
Expected
The suite is green on main, and the check names the four sets that are now offered (and keeps failing if a set
gains or loses its Mac words).
- Dominant language
- C++
- Stars
- 11.6k
- Forks
- 1k
- PR merge metrics
- No merged PRs in 30d
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 62/100
Niko1221/Strata#5138 · 2 comments ·
Maintainers usually reply within 3 days
-
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
Niko1221/Strata#5086 · 1 comment ·
Maintainers usually reply within 3 days
-
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
Maintainers usually reply within 3 days
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 3 days
-
[Other] Control hangs in platform.system() when Windows WMI is stuck: «Strata Control did not answer within 60 s (models list)», no GPU, nothing loadsPossibly taken A pull request linked to this issue is open or already merged. Open
Difficulty 2/5 1-3 hours Newbie friendliness 62/100
Maintainers usually reply within 3 days
Similar issues
-
bug libcudf
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
Maintainers usually reply within 2 days
-
level/task module/cloud monitoring type/maintenance
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
LadybirdBrowser/ladybird#13809 ·
Maintainers usually reply within 1 day
-
bug module:nodes tests
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
rocketride-org/rocketride-server#3020 ·
Maintainers usually reply within 2 days
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 9 days