unsloth/scripts
Daniel Han 15d951a0f5 Merge origin/main into fix/torchcodec-torch211
Two conflicts, both resolved by keeping each side's intent.

studio/backend/requirements/extras-no-deps.txt: main pinned the audio extras
to exact versions; this branch removed the flat torchcodec pin because the
wheel is torch-ABI specific and a requirements file cannot branch on the
installed torch minor. Kept main's exact pins for every other entry and kept
the torchcodec line removed, with the comment that points at
_select_torchcodec_spec.

studio/install_python_stack.py: both sides added at the same anchor. Main
replaced the standalone torch probe with a memoized _probe_torch_runtime and
kept _probe_installed_torch_version as a thin wrapper over it; this branch
added the torchcodec spec table and _select_torchcodec_spec just above. Kept
both blocks. Step 13b keeps calling _probe_installed_torch_version, so it now
shares main's single probe instead of spawning its own.

base_total stays at 13/14: main did not add or drop a progress step, and the
three torchcodec sites are mutually exclusive, so exactly one extra step runs.

tests/studio/install/test_diffusers_pin.py: the ordering check scanned the raw
install_python_stack.py source for requirements filenames, so the torchcodec
step's comment explaining why the spec cannot live in extras-no-deps.txt read
as an install after the diffusers pin and failed the test. The check is about
where `uv pip install -r` runs, so it now scans with comments blanked out and
offsets preserved. A real later install still trips it.
2026-08-16 21:09:58 +00:00
..
data Make the Colab oracle tripwire able to fire, and stop blaming start.py for a hung agent CLI (#7838) 2026-08-04 07:42:42 -07:00
build_prequant_checkpoint.py Studio: run MiniMax-H3's INT8 denoiser from the ConvRot checkpoint (#8293) 2026-08-10 01:00:16 -07:00
build_te_prequant_checkpoint.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
build_whisper_cpp.sh Studio: add local speech-to-text dictation engine (#7095) 2026-07-23 01:39:03 -07:00
check_frontend_dep_removal.py Reduce and tighten code comments and docstrings repo-wide (#6095) 2026-06-08 23:09:51 -07:00
check_new_install_scripts.py Reduce and tighten code comments and docstrings repo-wide (#6095) 2026-06-08 23:09:51 -07:00
compare_engines.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
compile_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
diffusion_bench.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
diffusion_quality.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
enforce_kwargs_spacing.py Reduce and tighten code comments and docstrings repo-wide (#6095) 2026-06-08 23:09:51 -07:00
fbcache_flux_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
fp8_overflow_check.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
image_speedmem_bench.py Report the precision actually in use, and refuse an explicit one we cannot honor (#8165) 2026-08-09 01:15:13 -07:00
install_gemma4_mlx.sh Update Install Scripts (#5968) 2026-06-03 05:39:42 -07:00
install_qwen3_6_mlx.sh Studio: drop the mlx-lm 0.31.3 exclusion so current mlx-vlm resolves (#7061) 2026-08-07 02:41:04 -03:00
install_rocm_wsl_strixhalo.sh Pin the ROCm-on-WSL bootstrap to immutable refs (#8540) 2026-08-12 06:37:05 -07:00
int8_linear_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
leverage_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
lint_no_parallel_clamp.py Studio: stop MTP forcing llama-server to a single parallel slot (#8172) 2026-08-08 21:09:58 -07:00
lint_workflow_triggers.py Harden the workflow-trigger lint: scan .yaml, and host it outside the workflow it audits (#8545) 2026-08-12 09:55:26 -07:00
lockfile_supply_chain_audit.py security: lockfile audit must block non-registry sources and missing integrity by default (#8541) 2026-08-12 05:36:17 -07:00
make_dmg_background.py Start the bottom taper under the icon so the disc still reads round (#8321) 2026-08-10 01:06:59 -07:00
notebook_to_python.py Reduce and tighten code comments and docstrings repo-wide (#6095) 2026-06-08 23:09:51 -07:00
notebook_validator.py Merge branch 'main' into fix/torchcodec-torch211 2026-08-10 03:47:36 +00:00
nvfp4_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
nvfp4_t211_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
online_tokenization_ab.py Studio: tokenize the dataset online for plain-text single-pass runs (#8960) 2026-08-16 04:55:53 -07:00
perf_levers_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
perf_verify.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
prequant_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
profile_startup.py Measure where Studio's startup time actually goes (#7553) 2026-07-28 22:24:34 -07:00
quant_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
run_ruff_format.py Reduce and tighten code comments and docstrings repo-wide (#6095) 2026-06-08 23:09:51 -07:00
scan_npm_packages.py Replace standalone Studio wording with Unsloth (#7221) 2026-07-19 00:47:04 -07:00
scan_npm_packages_baseline.json scan_packages: key baseline on matched-code hash so payloads in baselined files are not auto-suppressed (#6552) 2026-07-01 04:03:59 -07:00
scan_packages.py Fix CI on main: stale test doubles, a stale router stub, and two source defects (#8956) 2026-08-16 02:02:47 -07:00
scan_packages_baseline.json Fix CI on main: stale test doubles, a stale router stub, and two source defects (#8956) 2026-08-16 02:02:47 -07:00
sd_cpp_smoke.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
sdpa_mask_backend_probe.py Diffusion: replace the attention perf claims with re-measured numbers (#8021) 2026-08-06 05:47:31 -07:00
sparse_accum_probe.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
stamp_studio_release.py Replace standalone Studio wording with Unsloth (#7221) 2026-07-19 00:47:04 -07:00
sync_allow_scripts_pins.py Studio: auto-sync allowScripts pins after dependency bumps (#6136) 2026-06-10 02:35:37 -07:00
uninstall.ps1 Windows: stop depending on the generated unsloth.exe console script (#8592) 2026-08-13 07:54:51 -07:00
uninstall.sh Studio: install sd.cpp under the Studio home, not beside it (#8226) 2026-08-10 07:03:20 -07:00
verify_comment_only_diff.py Reduce and tighten code comments and docstrings repo-wide (#6095) 2026-06-08 23:09:51 -07:00
verify_import_hoist.py feat(studio): rework train page setup flow (#7633) 2026-08-06 04:26:05 -07:00
verify_prequant_backend.py Studio: add image generation, editing workflows and LoRA training with Unsloth GGUFs (#6763) 2026-08-04 08:11:01 -07:00
video_quality.py Video: make MiniMax-H3's Diffusers path fast by default (#8320) 2026-08-10 04:31:09 -07:00
virustotal_scan.py Reduce antivirus false positives in the desktop installers (#8586) 2026-08-13 07:02:18 -07:00