Tashdid Khan e1ec301db3 Use NORMAL_VRAM instead of SHARED for MPS memory management ... On Apple Silicon, SHARED vram state is treated like HIGH_VRAM in several code paths, which keeps all models loaded on the GPU device. Since MPS uses unified memory (GPU memory IS system memory), this starves the system of memory for inference activations when running large models (e.g. Wan 2.2 14B I2V loads 13.6GB model + 21.6GB text encoder = 35GB, leaving only 13GB for activations on a 48GB system). This caused GPU hangs and kernel panics on macOS when generating video with 14B parameter models. Changes: - Set MPS vram_state to NORMAL_VRAM instead of SHARED, enabling smart memory management to offload unused models during inference - Remove SHARED from unet_inital_load_device HIGH_VRAM fast path, so model loading respects available memory on MPS - Remove MPS shortcut in text_encoder_initial_device that forced the text encoder to stay on GPU regardless of memory pressure On unified memory, "offloading to CPU" is nearly free (same physical memory pool, just a device pointer change), so NORMAL_VRAM behavior is strictly better — it lets ComfyUI intelligently manage what stays on the GPU device based on available memory. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>		2026-02-09 23:50:46 -05:00
..
audio_encoders	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
cldm	Add better error message for common error. (#10846)	2025-11-23 04:55:22 -05:00
comfy_types	Add support for dev-only nodes. (#12106)	2026-01-27 13:03:29 -08:00
extra_samplers	Uni pc sampler now works with audio and video models.	2025-01-18 05:27:58 -05:00
image_encoders	Add Hunyuan 3D 2.1 Support (#8714)	2025-09-04 20:36:20 -04:00
k_diffusion	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
ldm	Some fixes to previous pr. (#12339)	2026-02-06 20:14:52 -05:00
sd1_tokenizer	Silence clip tokenizer warning. (#8934)	2025-07-16 14:42:07 -04:00
t2i_adapter
taesd	Support LTX2 tiny vae (taeltx_2) (#11929)	2026-01-21 23:03:51 -05:00
text_encoders	Disable prompt weights for ltxv2. (#12354)	2026-02-07 19:16:28 -05:00
weight_adapter	[Weight-adapter/Trainer] Bypass forward mode in Weight adapter system (#11958)	2026-01-24 22:56:22 -05:00
checkpoint_pickle.py
cli_args.py	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
clip_config_bigg.json	Fix potential issue with non clip text embeddings.	2024-07-30 14:41:13 -04:00
clip_model.py	Support the siglip 2 naflex model as a clip vision model. (#11831)	2026-01-12 17:05:54 -05:00
clip_vision_config_g.json
clip_vision_config_h.json
clip_vision_config_vitl_336_llava.json	Support llava clip vision model.	2025-03-06 00:24:43 -05:00
clip_vision_config_vitl_336.json
clip_vision_config_vitl.json
clip_vision_siglip2_base_naflex.json	Support the siglip 2 naflex model as a clip vision model. (#11831)	2026-01-12 17:05:54 -05:00
clip_vision_siglip_384.json	Support new flux model variants.	2024-11-21 08:38:23 -05:00
clip_vision_siglip_512.json	Support 512 siglip model.	2025-04-05 07:01:01 -04:00
clip_vision.py	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
conds.py	Add some warnings and prevent crash when cond devices don't match. (#9169)	2025-08-04 04:20:12 -04:00
context_windows.py	Add handling for vace_context in context windows (#11386)	2025-12-30 14:40:42 -08:00
controlnet.py	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
diffusers_convert.py	Remove useless code.	2025-01-24 06:15:54 -05:00
diffusers_load.py	load_unet -> load_diffusion_model with a model_options argument.	2024-08-12 23:20:57 -04:00
float.py	Optimize nvfp4 lora applying. (#11866)	2026-01-14 00:49:38 -05:00
gligen.py	Remove some useless code. (#8812)	2025-07-06 07:07:39 -04:00
hooks.py	New Year ruff cleanup. (#11595)	2026-01-01 22:06:14 -05:00
latent_formats.py	Basic support for the ace step 1.5 model. (#12237)	2026-02-03 00:06:18 -05:00
lora_convert.py	Implement the USO subject identity lora. (#9674)	2025-09-01 18:54:02 -04:00
lora.py	Support ace step 1.5 base model loras. (#12252)	2026-02-03 13:54:23 -05:00
memory_management.py	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
model_base.py	Pad ace step 1.5 ref audio if not long enough. (#12341)	2026-02-07 00:02:11 -05:00
model_detection.py	Basic support for the ace step 1.5 model. (#12237)	2026-02-03 00:06:18 -05:00
model_management.py	Use NORMAL_VRAM instead of SHARED for MPS memory management	2026-02-09 23:50:46 -05:00
model_patcher.py	mp: Fix checkpoint saving (#12268)	2026-02-04 02:08:45 -05:00
model_sampling.py	Refactor model sampling sigmas code. (#10250)	2025-10-08 17:49:02 -04:00
nested_tensor.py	WIP way to support multi multi dimensional latents. (#10456)	2025-10-23 21:21:14 -04:00
ops.py	Try to fix ace text encoder slowness on some configs. (#12290)	2026-02-04 19:37:05 -05:00
options.py
patcher_extension.py	Fix order of inputs nested merge_nested_dicts (#10362)	2025-10-15 16:47:26 -07:00
pinned_memory.py	fix pinning with model defined dtype (#12208)	2026-02-01 08:42:32 -08:00
pixel_space_convert.py	Changes to the previous radiance commit. (#9851)	2025-09-13 18:03:34 -04:00
quant_ops.py	Optimize nvfp4 lora applying. (#11866)	2026-01-14 00:49:38 -05:00
rmsnorm.py	Add warning when using old pytorch. (#9347)	2025-08-15 00:22:26 -04:00
sample.py	Make regular empty latent node work properly on flux 2 variants. (#12050)	2026-01-23 19:50:48 -05:00
sampler_helpers.py	skip_load_model -> force_full_load (#11390)	2025-12-17 23:29:32 -05:00
samplers.py	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00
sd1_clip_config.json	Fix potential issue with non clip text embeddings.	2024-07-30 14:41:13 -04:00
sd1_clip.py	Basic support for the ace step 1.5 model. (#12237)	2026-02-03 00:06:18 -05:00
sd.py	Add VAE tiled decode node for audio. (#12299)	2026-02-05 01:12:04 -05:00
sdxl_clip.py	Add a T5TokenizerOptions node to set options for the T5 tokenizer. (#7803)	2025-04-25 19:36:00 -04:00
supported_models_base.py	Fix some custom nodes. (#11134)	2025-12-05 18:25:31 -05:00
supported_models.py	Some fixes to previous pr. (#12339)	2026-02-06 20:14:52 -05:00
utils.py	utils: safetensors: dont slice data on torch level (#12266)	2026-02-04 01:48:47 -05:00
windows.py	Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)	2026-02-01 01:01:11 -05:00