5340 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Talmaj Marinc	aefc61f42d	Refactor and rename nodes to from DepthAnyting3 to use DA3 naming convention.	2026-05-26 14:30:08 +02:00
Talmaj Marinc	15c096aa16	Fix confidence use in DA3ToMesh by using normalization, fix extrisinc usage.	2026-05-26 14:26:26 +02:00
Talmaj Marinc	2dd9d96d4a	Fix extrinsic quaternions by normalizing them to unit quaternion.	2026-05-26 14:25:18 +02:00
Talmaj Marinc	2ed1f36471	Add DA3GeometryToMesh and DA3GeometryToPointCloud	2026-05-26 13:47:35 +02:00
Talmaj Marinc	7cb2394630	Improve and correct error logging for mono/multi-view.	2026-05-26 13:42:35 +02:00
Talmaj Marinc	ddb8739963	Improve tooltips.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	e3a74d1696	Fixup last few naming issues.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	359da6d0b4	Revert support for DA3Geometry for MoGeRender.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	b65bdc3737	Move turbo to colormap.py and add it to DepthAnything3Render.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	d66f385502	Refactor DepthAnything3 into DepthAnything3Inference and DepthAnything3Render.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	22982da481	Refactor to a single socket ouptut for DA3 compatible with MoGeRender.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	9e30a0b56c	Add DA3ModelType to LoadDepthAnything3.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	81c8afb36d	Improve tooltip for normalization.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	102704f9fb	Apply some PR comments.	2026-05-26 12:52:14 +02:00
Talmaj	ccdc6517fd	Fix DA3 model loader category. ... Co-authored-by: Alexis Rolland <alexisrolland@hotmail.com>	2026-05-26 12:52:14 +02:00
Talmaj Marinc	33c0421153	Fix node loading from geometry_estimation folder.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	50729162f4	Modify apply_sky_clip default to False.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	d6aad2f8c7	Fix no clamping when normalization is raw and refactor reusable code to _apply_sky_clip and _depth_to_image	2026-05-26 12:52:14 +02:00
Talmaj Marinc	67278c5851	Make LoadDepthAnything3 look for models in geometry_estimation.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	864249d3c1	Normalize confidence output.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	d7d149b754	Improve the toggles text.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	47d3d90380	Cleanup of comments.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	33235ab099	Create a unified DepthAnything3 node.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	a9a993130e	Remove unused code.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	c3c9ce39f9	Update DA3 to use dino2.get_intermediate_layers_da3	2026-05-26 12:52:14 +02:00
Talmaj Marinc	fb19982410	Merge interpolate_pos_encoding method in dinov2	2026-05-26 12:52:14 +02:00
Talmaj Marinc	19c25c8cbd	Fix ray_pose for non float32 dtypes CORE-135	2026-05-26 12:52:14 +02:00
Talmaj Marinc	33701de3a8	Add support for pose estimation CORE-135.	2026-05-26 12:52:14 +02:00
Talmaj Marinc	82e4db5d4a	Refactor custom da dinov2 to image_encoders/dino2	2026-05-26 12:52:14 +02:00
Talmaj Marinc	1ffa010952	Modify upscaling to be closer to original CORE-135	2026-05-26 12:52:14 +02:00
Talmaj Marinc	9b19fab3da	Initial commit CORE-135	2026-05-26 12:52:13 +02:00
Jukka Seppänen	f9f54cae42	Lens: some cleanup (#14112) ... * Lens: remove redundant memory optimization	2026-05-26 10:32:53 +03:00
Jukka Seppänen	41812fa0ac	feat: Microsoft Lens support (CORE-248) (#14077)	2026-05-25 23:01:51 -07:00
Ivan Zorin	57414dadfe	fix: cross-attention AdaLN scale, shift, sigma parameters calculation (#14097)	2026-05-25 20:07:09 -07:00
Jedrzej Kosinski	88956e77af	multigpu: use unet_manual_cast for SelectModelDevice compute dtype (#14108)	2026-05-25 20:03:37 -07:00
comfyanonymous	da49b7d0b6	Remove useless annotations imports. (#14105)	2026-05-25 19:23:29 -07:00
Jedrzej Kosinski	0a2dd86e78	MultiGPU Work Units For Accelerated Sampling (CORE-184) (#7063)	2026-05-25 18:26:40 -07:00
Daxiong (Lin)	04879a8113	Add new open-source model and built-in tool blueprints (#13980)	2026-05-25 12:25:16 -07:00
Matt Miller	6de7fc063b	Emit hash alongside asset_hash on all Asset responses (#13739) ... * Emit `hash` alongside `asset_hash` on all Asset responses Add a `hash` field to the Asset response schema that carries the same value as the existing `asset_hash` field. Both fields are now populated in _build_asset_response, so every Asset-returning endpoint (GET, POST, PUT) includes both. No existing fields are removed. Tests updated to assert both fields. Co-authored-by: Matt Miller <MillerMedia@users.noreply.github.com> * Tighten hash field tests and DRY response builder - Extract assert_hash_fields_consistent() helper that verifies presence parity and value equality, replacing body.get()-based assertions that treated missing keys and explicit nulls identically. - Conftest seeded_asset fixture and seed-asset list assertions now check key absence directly, so a regression that surfaces null fields would be caught (validates exclude_none behavior). - DRY duplicate hash expression in _build_asset_response. - Add list-endpoint coverage asserting hash is present and consistent on populated assets. - Add schema-level test asserting AssetCreated inherits the hash field from Asset, guarding against future inheritance drift. --------- Co-authored-by: Matt Miller <MillerMedia@users.noreply.github.com> Co-authored-by: guill <jacob.e.segal@gmail.com>	2026-05-25 11:21:35 -07:00
Daxiong (Lin)	a4141a0f5a	chore: update embedded docs to v0.5.1 (#14101)	2026-05-26 01:57:18 +08:00
comfyanonymous	0077d78cbf	Save Image advanced node (CORE-32) (#13850)	2026-05-24 23:01:34 -04:00
Talmaj	63bcaec5d1	Add colored logs (#14036)	2026-05-25 10:00:55 +08:00
rattus	b30e980a20	cache-ram: lower thresholds (#14089) ... Use the RAM right up to the wire as the community is bit accustomed too. This trades off headroom for the case where large chunky intermediates arrive and potenitally hits pagefile/swap, but a lot of people have "it just fits" workflows out there, so strike a compromise with 75->90%. Disable the incative cache for all but the very high RAM users.	2026-05-24 15:26:50 -07:00
rattus	39f963b4b0	mark loads to pins as cold immediately (#14088) ... This does the posix_fadvise to kick pins out of the disk cache (to avoid a double copy in RAM).	2026-05-24 15:25:59 -07:00
Matt Miller	ea62dc11c9	openapi: fix invalid BillingStatus schema (object + enum hybrid) (#14071)	2026-05-24 10:58:35 +08:00
Robin Huang	32a7092c52	fix: correct description of where compiled FE files live (#14013)	2026-05-24 10:48:31 +08:00
comfyanonymous	08d809d128	Fix --use-flash-attention ignored when xformers installed. (#14083)	2026-05-23 17:44:28 -07:00
Comfy Org PR Bot	0af123022d	Bump comfyui-frontend-package to 1.44.19 (#14074)	2026-05-24 08:27:52 +08:00
comfyanonymous	d80fcafee7	Remove dead code. (#14072)	2026-05-22 19:56:36 -07:00
Matt Miller	187442cca4	openapi: add enum values + FeedbackRequest schema for cloud cutover (PR E) (#14070) ... * openapi: add enum values + FeedbackRequest schema for cloud cutover (PR E) Adds missing cloud-runtime enum values to vendor schemas that the cloud runtime emits but vendor declared as plain strings. Changes: - JobEntry.status: enum [pending, in_progress, completed, failed, cancelled] - JobDetailResponse.status: same enum - BillingStatus: enum [awaiting_payment_method, pending_payment, paid, payment_failed, inactive] - FeedbackRequest schema added (with type enum) - /api/feedback POST: requestBody now $refs FeedbackRequest All cloud-runtime-emitted; no impact on OSS-local semantics. Identified via Comfy-Org/cloud's TestCutoverSafe gate (BE-1106) as the remaining schema-level divergences after PRs A-D landed and got synced. * openapi: add type enum to Workspace schema (cutover follow-up) Cloud's Workspace runtime shape includes a 'type' field with enum [personal, team] that vendor's Workspace was missing. Cloud handlers reference the generated ingest.WorkspaceType Go enum. Same kind of surgical addition as JobEntry.status / BillingStatus / JobDetailResponse.status in this PR — adds cloud-runtime field to existing vendor schema.	2026-05-22 18:23:22 -07:00