Aryan Kumar fd9f06654e [bugfix]: apply pika_resolution_tier when a request omits its size ...

The sampling param and presets fill 768x1344 before input preparation, so
an exact p768 bucket always won and the configured tier never applied.
Resolve the omitted canvas in apply_request_constraints, where explicit
request fields are still distinguishable; explicit sizes keep their tier
and one-sided sizes are rejected.

2026-10-29 18:51:52 -07:00

..

backend/vmoba

[Backend][Vmoba] Add implementation of VMoba (#778)

2025-09-08 23:53:25 -07:00

models

[feat]: add Pika H3 Flex inference support

2026-10-29 17:37:52 -07:00

pipelines

[bugfix]: apply pika_resolution_tier when a request omits its size

2026-10-29 18:51:52 -07:00

__init__.py

[Feature] Remove V1 folder (#642)

2025-07-24 22:43:12 -07:00

AGENTS.md

[refactor] Simplify Wan sampling and tests (#1825)

2026-09-07 15:57:19 -07:00

configs.py

[feat] Add end-to-end MMAudio V2A training, inference, and evaluation (#1648)

2026-10-07 04:25:00 -07:00

fasthunyuan_t2v.json

[Misc] Remove STA (#1124)

2026-02-23 15:14:42 -08:00

utils.py

[feat] pre-commit support 120 col num (#1167)

2026-03-11 20:19:30 -07:00

wan_1.3B_t2v_pipeline.json

[perf] Default Wan VAE decode to bf16 (lossless, faster) (#1472)

2026-06-26 12:42:29 -07:00

wan_14B_i2v_480p_pipeline.json

[perf] Default Wan VAE decode to bf16 (lossless, faster) (#1472)

2026-06-26 12:42:29 -07:00