nodes/hao-ai-lab-FastVideo
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[bugfix]: apply the LTX-2 refine LoRA to the refine transformer

Browse Source 
The from_pretrained keyword ltx2_refine_lora_path set
pipeline.components.lora_path, which reaches FastVideoArgs.lora_path and loads
as the main LoRA on the stage-1 transformer. The refine stage reads
FastVideoArgs.ltx2_refine_lora_path, which stayed unset. The conversion also
turned an empty string into None. An empty string is how callers such as the
Dreamverse GPU pool disable the refine LoRA, and None makes LTX2Pipeline load
the checkpoint default instead: fastvideo_refine_lora_path in model_index.json
is FastVideo/LTX2-Distilled-LoRA for the LTX-2 Diffusers checkpoints.

The refine LoRA gets its own field, pipeline.ltx2.refine.lora_path, whose
flat name is ltx2_refine_lora_path. None keeps the checkpoint default, and an
empty string disables the refine LoRA. The Dreamverse contract test, the
Dynamo contract example, and the parity inventory point at the field.

Golden changes: kwargs/ltx2_refine_lora_path now records the path on
ltx2_refine_lora_path instead of lora_path, and the added case
kwargs/ltx2_refine_lora_disabled pins the empty string.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012M91wnVFmPEvJ9h39r5BH7
This commit is contained in:
Davids048
2026-10-02 18:29:37 +00:00
co-authored by Claude Opus 5.5
parent 991fa391ff
commit c7c2d77fae
12 changed files with 382 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/inference/basic/basic_ltx2_3_distilled_i2v_typed.py
+1 -1
View File
@@ -56,7 +56,7 @@ Typed-API mapping (legacy kwarg ↔ typed field)
- ``ltx2_vae_tiling`` ↔ ``pipeline.vae_tiling``
- ``ltx2_refine_enabled`` ↔ ``pipeline.preset_overrides["refine"]["enabled"]``
- ``ltx2_refine_upsampler_path`` ↔ ``pipeline.components.upsampler_weights``
- ``ltx2_refine_lora_path`` ↔ ``pipeline.components.lora_path``
- ``ltx2_refine_lora_path`` ↔ ``pipeline.ltx2.refine.lora_path``
- ``ltx2_refine_num_inference_steps`` ↔ ``pipeline.preset_overrides["refine"]["num_inference_steps"]``
- ``ltx2_refine_guidance_scale`` ↔ ``pipeline.preset_overrides["refine"]["guidance_scale"]``
- ``ltx2_refine_add_noise`` ↔ ``pipeline.preset_overrides["refine"]["add_noise"]``