v0.3.18

What's Changed

• Improve Wan performance by allowing batching and fix issue with long prompts.
• Support Cambricon MLU by @BiologicalExplosion in #6964

New Contributors

• @BiologicalExplosion made their first contribution in #6964

Full Changelog: v0.3.17...v0.3.18

v0.3.17

• Fix artifacts when inferencing wan with fp8 weights.

Full Changelog: v0.3.16...v0.3.17

v0.3.16

What's Changed

• Initial support for new Wan 2.1 video models
• Update installation documentation to include desktop + cli. by @robinjhuang in #6899
• Update README.md by @yoland68 in #6960

Full Changelog: v0.3.15...v0.3.16

v0.3.15

What's Changed

• Some improvements for AMD ROCm.
• [Cleanup] Remove unused GET internal/files API endpoint by @huchenlei in #6714
• Frontend Update: v1.8.14 by @comfy-pr-bot in #6724
• fix a bug in the attn_masked redux code when using weight=1.0 by @Slickytail in #6721
• Fix another small bug in attention_bias redux by @Slickytail in #6737
• Allow FP16 accumulation with --fast by @catboxanon in #6453
• res_multistep: Fix cfgpp and add ancestral samplers by @pamparamm in #6731
• Fix incorrect Content-Type for WebP images by @bananasss00 in #6752
• Fix for running via DirectML by @hisham-hchowdhu in #6542
• Fix the Issue with Inference on Ascend NPU by @zhoufan2956 in #6794
• Open yaml files with utf-8 encoding for extra_model_paths.yaml by @robinjhuang in #6807
• Frontend Update: v1.9.17 by @comfy-pr-bot in #6814
• remove light_intensity and fov from load3d by @jtydhr88 in #6742
• Frontend Update: v1.9.18 by @comfy-pr-bot in #6828
• support system prompt and cfg renorm in Lumina2 by @lzyhha in #6795
• Support loading and using SkyReels-V1-Hunyuan-I2V by @kijai in #6862
• Add Load Image Output node by @christian-byrne in #6790
• Fix Hunyuan unet config detection for some models. by @maedtb in #6877
• fixed: crash caused by outdated incompatible aiohttp dependency by @ltdrdata in #6841
• Fix link pointing to non-exisiting docs by @silveroxides in #6891
• Normalize extra_model_config.yaml paths to prevent duplicates. by @robinjhuang in #6885
• Add discord channel to support section of README. by @robinjhuang in #6900
• Update frontend release schedule in README. by @webfiltered in #6908
• Various fixes and improvements.

New Contributors

• @bananasss00 made their first contribution in #6752
• @hisham-hchowdhu made their first contribution in #6542
• @zhoufan2956 made their first contribution in #6794
• @lzyhha made their first contribution in #6795
• @maedtb made their first contribution in #6877

Full Changelog: v0.3.14...v0.3.15

v0.3.14

What's Changed

• Support for Lumina Image 2.0
• Frontend Update: v1.8.12 by @comfy-pr-bot in #6662
• Add node for preview 3d animation by @jtydhr88 in #6594
• add disable-compres-response-body cli args; add compress middleware; by @KarryCharon in #6672
• improve: better guide message for sageattention by @ltdrdata in #6634
• Frontend Update: v1.8.13 by @comfy-pr-bot in #6682
• Allow searching for .pt2 archives by @Slickytail in #6689

Full Changelog: v0.3.13...v0.3.14

v0.3.13

What's Changed

• Update pytorch in standalone package to pytorch 2.6 cuda 12.6
• Nvidia Blackwell/50 series support, see: this discussion thread
• Drop support for python 3.8
• Better error messages for a few common errors.
• Fix uni_pc samplers not working on audio and video models.
• Add FluxDisableGuidance node to disable the guidance embed on Flux and Hunyuan video models.
• PromptServer: Return 400 for empty filename param by @catboxanon in #6504
• Remove comfy.samplers self-import by @catboxanon in #6506
• Frontend Update: v1.7.14 by @comfy-pr-bot in #6522
• Use torch.special.expm1 by @kit1980 in #6388
• Implement gradient-estimation sampler by @chaObserv in #6554
• [i18n] Add /i18n endpoint to provide all custom node translations by @huchenlei in #6558
• Add utils/ to web server developer codeowner by @huchenlei in #6570
• remove some params from load 3d node by @jtydhr88 in #6436
• [Cleanup] Remove unused function lcm in conds.py by @huchenlei in #6572
• Remove redundant code from folder_paths by @webfiltered in #6576
• Convert latents_ubyte to 8-bit unsigned int before converting to CPU by @shenanigansd in #6300
• Allow changing folder_paths.base_path via command line argument. by @webfiltered in #6600

New Contributors

• @comfy-pr-bot made their first contribution in #6522
• @kit1980 made their first contribution in #6388
• @shenanigansd made their first contribution in #6300

Full Changelog: v0.3.12...v0.3.13

v0.3.12

What's Changed

• Nvidia Cosmos 7B and 14B: text to video and image to video diffusion model support.
• New sampler: res_multistep
• ckpt/pt/etc.. files are now always loaded safely on pytorch 2.4 and above.
• Fix some cases of ancestral samplers not being deterministic.
• Support ascend npu by @ji-huazhong in #5436
• Add option to log non-error output to stdout by @webfiltered in #6243
• serve workflow templates from custom_nodes by @bezo97 in #6193
• Remove duplicate calls to INPUT_TYPES by @catboxanon in #6249
• Fix Hook Keyframe 'guarantee_steps' behavior and add 'sigmas' by @Kosinkadink in #6273
• Add kl_optimal scheduler by @blepping in #6206
• (fix): "verbose" argument by @bigcat88 in #6289
• Fix custom node type-hinting examples by @webfiltered in #6281
• Add missing model_options param to finalize_default_conds call by @Kosinkadink in #6296
• Fix unknown scheduler error handling in calculate_sigmas function by @blepping in #6280
• Fix temporal tiling for Tiled VAE decoder, remove redundant tiles. by @kvochko in #6306
• Update web content to release v1.6.14 by @huchenlei in #6312
• add fov and mask for load 3d node by @jtydhr88 in #6308
• Update web content to release v1.6.15 by @huchenlei in #6324
• Update web content to release v1.6.16 by @huchenlei in #6335
• Update web content to release v1.6.17 by @huchenlei in #6337
• Add update-frontend github action by @huchenlei in #6336
• Update CODEOWNERS by @yoland68 in #6338
• In inner_sample, change "sigmas" to "sample_sigmas" by @Kosinkadink in #6360
• Frontend Update: v1.6.18 by @huchenlei in #6368
• Document get_attr and get_model_object by @huchenlei in #6357
• fixed: robust loading comfy.settings.json by @ltdrdata in #6383
• Add pyproject.toml by @huchenlei in #6386
• Hooks Part 2 - TransformerOptionsHook and AdditionalModelsHook by @Kosinkadink in #6377
• Merge ruff.toml into pyproject.toml by @huchenlei in #6431
• (fix): load_extra_path_config: relative path not converted to a full path by @bigcat88 in #6395
• Rewrite res_multistep sampler and implement res_multistep_cfg_pp sampler by @pamparamm in #6462
• Add SetFirstSigma node by @catboxanon in #6459

New Contributors

• @ji-huazhong made their first contribution in #5436
• @bezo97 made their first contribution in #6193
• @kvochko made their first contribution in #6306

Full Changelog: v0.3.10...v0.3.12

v0.3.11

What's Changed

• Nvidia Cosmos 7B and 14B: text to video and image to video diffusion model support.
• New sampler: res_multistep
• ckpt/pt/etc.. files are now always loaded safely on pytorch 2.4 and above.
• Fix some cases of ancestral samplers not being deterministic.
• Support ascend npu by @ji-huazhong in #5436
• Add option to log non-error output to stdout by @webfiltered in #6243
• serve workflow templates from custom_nodes by @bezo97 in #6193
• Remove duplicate calls to INPUT_TYPES by @catboxanon in #6249
• Fix Hook Keyframe 'guarantee_steps' behavior and add 'sigmas' by @Kosinkadink in #6273
• Add kl_optimal scheduler by @blepping in #6206
• (fix): "verbose" argument by @bigcat88 in #6289
• Fix custom node type-hinting examples by @webfiltered in #6281
• Add missing model_options param to finalize_default_conds call by @Kosinkadink in #6296
• Fix unknown scheduler error handling in calculate_sigmas function by @blepping in #6280
• Fix temporal tiling for Tiled VAE decoder, remove redundant tiles. by @kvochko in #6306
• Update web content to release v1.6.14 by @huchenlei in #6312
• add fov and mask for load 3d node by @jtydhr88 in #6308
• Update web content to release v1.6.15 by @huchenlei in #6324
• Update web content to release v1.6.16 by @huchenlei in #6335
• Update web content to release v1.6.17 by @huchenlei in #6337
• Add update-frontend github action by @huchenlei in #6336
• Update CODEOWNERS by @yoland68 in #6338
• In inner_sample, change "sigmas" to "sample_sigmas" by @Kosinkadink in #6360
• Frontend Update: v1.6.18 by @huchenlei in #6368
• Document get_attr and get_model_object by @huchenlei in #6357
• fixed: robust loading comfy.settings.json by @ltdrdata in #6383
• Add pyproject.toml by @huchenlei in #6386
• Hooks Part 2 - TransformerOptionsHook and AdditionalModelsHook by @Kosinkadink in #6377
• Merge ruff.toml into pyproject.toml by @huchenlei in #6431
• (fix): load_extra_path_config: relative path not converted to a full path by @bigcat88 in #6395
• Rewrite res_multistep sampler and implement res_multistep_cfg_pp sampler by @pamparamm in #6462
• Add SetFirstSigma node by @catboxanon in #6459

New Contributors

• @ji-huazhong made their first contribution in #5436
• @bezo97 made their first contribution in #6193
• @kvochko made their first contribution in #6306

Full Changelog: v0.3.10...v0.3.11

v0.3.10

What's Changed

• Add temporal tiling support to video VAEs exposed through the VAE Decode (tiled) or VAE Encode (tiled) nodes, this makes 8GB vram GPUs viable for hunyuan video.
• Add PixArt model support by @city96 in #6055
• Replace print with logging by @huchenlei in #6138
• Adding upstream pytorch installation instruction for Intel GPUs to README.md by @qiacheng in #6069
• remove minimum step count for AYS by @TechnoByteJS in #6137
• fixed a bug in ldm/pixart/blocks.py by @zhangp365 in #6158
• Add oneAPI device selector for xpu and some other changes. by @simonlui in #6112
• [Bug] Use raw dir name to serve static web content by @huchenlei in #6107
• Enable External Event Loop Integration for ComfyUI [refactor] by @bigcat88 in #6114
• VAE calculations are now done in bf16 by default on Mac for lower memory usage.

New Contributors

• @qiacheng made their first contribution in #6069
• @TechnoByteJS made their first contribution in #6137
• @zhangp365 made their first contribution in #6158

Full Changelog: v0.3.9...v0.3.10

v0.3.9

What's Changed

• LTXV 0.9.1 support (new VAE architecture)
• Tiled VAE encode node now works with video VAEs
• Fix a memory issue that lead to OOM when using a partially loaded model multiple times in a row.
• Support preview images embedded in safetensors metadata by @catboxanon in #6119

Full Changelog: v0.3.8...v0.3.9