Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / THUDM/CogVideo issues and pull requests

#709 - [WIP] Implement DDIM Inversion for CogVideoX

Pull Request - State: open - Opened by LittleNyima 3 days ago

#707 - How to load full finetune for inference?

Issue - State: open - Opened by pldlgb 5 days ago - 4 comments
Labels: good first issue

#706 - Can I2V model be finetuned with less than 49 frames?

Issue - State: closed - Opened by daoyuan98 9 days ago - 2 comments

#705 - Question about CogVideoX 5B's resolution

Issue - State: open - Opened by wyyfffff 9 days ago - 1 comment

#704 - multi-gpu inference

Issue - State: open - Opened by cdfan0627 10 days ago - 4 comments

#702 - about preprocess_video_with_resize

Issue - State: open - Opened by renrenzsbbb 10 days ago

#701 - CogVideoX2

Issue - State: open - Opened by cdfan0627 11 days ago - 1 comment

#700 - Get Type Error when using multiprocessing

Issue - State: closed - Opened by hey0chen 11 days ago - 1 comment

#699 - cogvideox-i2v-5b differsers inference error?

Issue - State: closed - Opened by HuazhangHu 12 days ago - 1 comment

#698 - multi-gpu validation

Issue - State: open - Opened by renrenzsbbb 14 days ago

#697 - Reference error in the section 2.1?

Issue - State: open - Opened by gobigrassland 15 days ago

#696 - modelscope页面打开失败

Issue - State: open - Opened by co63oc 15 days ago

#695 - XDIT multi GPUs inference code runs OOM with 24 frames

Issue - State: open - Opened by danielajisafe 17 days ago - 2 comments

#694 - Question: concat_image and ops

Issue - State: open - Opened by yjhong89 17 days ago - 1 comment
Labels: sat

#693 - Different fine-tuning modules

Issue - State: open - Opened by yjhong89 17 days ago - 1 comment

#692 - decord is not compatible with python 3.9+ and macos

Issue - State: closed - Opened by Tayapanic 23 days ago - 1 comment

#691 - does it support multiple gpu

Issue - State: open - Opened by BASSEM45325 24 days ago - 4 comments

#689 - Video Reconstruction based on Inversion

Issue - State: open - Opened by JonathanLi19 25 days ago - 3 comments

#688 - How to load the sft finetuned model based on the diffusers version.

Issue - State: open - Opened by zyhbili 25 days ago - 2 comments

#687 - cogvideo 1.0全参数 diffuser 微调代码

Issue - State: open - Opened by linwenzhao1 27 days ago - 4 comments

#686 - Attention Mask for T5 / T2V Attention

Issue - State: open - Opened by karan-dalal 28 days ago

#684 - update will suitable for uoload pretrained model

Issue - State: open - Opened by renrenzsbbb 30 days ago - 1 comment

#683 - CogVideoXDPMScheduler

Issue - State: closed - Opened by JonathanLi19 about 1 month ago

#682 - Synchronize two branches.

Pull Request - State: closed - Opened by zRzRzRzRzRzRzR about 1 month ago

#681 - training resolution support

Issue - State: closed - Opened by Yuqifan1117 about 1 month ago - 5 comments

#680 - CogVideoX-5B-I2V生成更長秒數影片問題

Issue - State: open - Opened by Labibibidu about 1 month ago - 12 comments

#679 - Issue about the latents frames number of CogVideoX1.5-5B

Issue - State: closed - Opened by onevfall about 1 month ago - 4 comments

#678 - docs: clarify frame number requirements for CogVideoX models

Pull Request - State: closed - Opened by OleehyO about 1 month ago

#677 - Cogvideox-5B-i2v inferences videos all mosaics

Issue - State: closed - Opened by yixuanli98 about 1 month ago - 2 comments

#676 - lateset Diffuser: Inference T2V meets patch_size_t error in CogVideoX-2B/5B

Issue - State: open - Opened by onevfall about 1 month ago - 2 comments

#675 - Training-free acceleration for CogVideoX

Issue - State: open - Opened by LiewFeng about 1 month ago

#674 - Issue in VAE compression

Issue - State: open - Opened by zhang-haojie about 1 month ago

#673 - CogVideoX1.5-5B-I2V 4090 24g Cannot run

Issue - State: open - Opened by hjj-lmx about 1 month ago - 8 comments

#672 - cli_demo.py an error occured

Issue - State: closed - Opened by ytm-01 about 1 month ago - 3 comments

#671 - TypeError: 'int' object is not subscriptable

Issue - State: closed - Opened by Jiro-zhang about 1 month ago - 1 comment
Labels: sat

#670 - ModuleNotFoundError: No module named 'moviepy.editor'

Issue - State: closed - Opened by Zdong104 about 1 month ago

#669 - CogVideoX1.5 t2v Diffusers Fine-tuning an error occurred

Issue - State: closed - Opened by ytm-01 about 1 month ago - 2 comments

#668 - Is it normal to generate a 20 minute session on 3090?

Issue - State: closed - Opened by hjj-lmx about 1 month ago - 1 comment

#667 - Getting following error on google colab

Issue - State: open - Opened by smktech9 about 1 month ago - 1 comment

#666 - Giving Transformers Layer cannot be '' error

Issue - State: open - Opened by smktech9 about 1 month ago - 2 comments

#664 - CogVideoX and CogvideoX1.5 video quality

Issue - State: open - Opened by cdfan0627 about 1 month ago - 2 comments

#663 - prepare_rotary_positional_embeddings training and inference are different

Issue - State: open - Opened by cdfan0627 about 1 month ago - 2 comments

#662 - Theoritically, what limits the number of frames the model can inferrence?

Issue - State: open - Opened by HWT-WalterHu about 1 month ago - 2 comments

#661 - Training video size

Issue - State: open - Opened by cdfan0627 about 1 month ago - 5 comments

#660 - latent padding training and inference different

Issue - State: open - Opened by cdfan0627 about 1 month ago - 14 comments

#659 - Reason for 49 frames (extra split for interpolation)

Issue - State: open - Opened by karan-dalal about 1 month ago - 4 comments

#658 - why 5b vae use a small latent scaling factor?

Issue - State: open - Opened by chengmengli06 about 1 month ago

#657 - fix bug of i2v finetune

Pull Request - State: closed - Opened by ZGCTroy about 1 month ago

#656 - fix potential bug of i2v

Pull Request - State: closed - Opened by ZGCTroy about 1 month ago

#655 - 模型转为diffusers有精度损失

Issue - State: open - Opened by linwenzhao1 about 1 month ago - 4 comments

#654 - Support SFT using ZeRO

Pull Request - State: closed - Opened by OleehyO about 1 month ago - 1 comment

#653 - 修改Sat的readme文件

Issue - State: closed - Opened by xiaolong-li1 about 1 month ago - 1 comment

#652 - Update code and requirements to support Moviepy v2

Pull Request - State: closed - Opened by erfanasgari21 about 1 month ago - 1 comment

#651 - Fix Moviepy compatibility issue and update README instructions

Pull Request - State: closed - Opened by erfanasgari21 about 1 month ago

#650 - Add image into the first frame

Issue - State: open - Opened by cdfan0627 about 1 month ago

#649 - Visual Video DIT Attention map

Issue - State: open - Opened by Passenger12138 about 1 month ago
Labels: good first issue

#648 - The video of I2V is a mosaic

Issue - State: closed - Opened by ZhiXinChasingLight about 1 month ago - 5 comments

#647 - finetune CogVideoX frames

Issue - State: open - Opened by cdfan0627 about 1 month ago - 15 comments

#646 - VAE training code

Issue - State: open - Opened by cdfan0627 about 1 month ago - 8 comments

#645 - CogVideoX1.5 LoRA Finetuning

Issue - State: closed - Opened by karan-dalal about 1 month ago - 3 comments

#644 - Windows system cannot install pip install - r requirements. txt

Issue - State: open - Opened by hjj-lmx about 2 months ago - 2 comments

#643 - Why does pad_last_frame() use last frame?

Issue - State: closed - Opened by karan-dalal about 2 months ago - 6 comments
Labels: sat

#642 - feat add & bugfix

Pull Request - State: closed - Opened by OleehyO about 2 months ago

#641 - v2v bugs when use CogVideoX1.5-5B

Issue - State: closed - Opened by YanZhuHotstar about 2 months ago - 3 comments

#640 - 关于3D full attention的问题

Issue - State: closed - Opened by Zhenhang-Li about 2 months ago - 1 comment

#639 - Multi-ID customization with CogvideoX 5B

Issue - State: open - Opened by feizc about 2 months ago - 1 comment
Labels: good first issue

#638 - Extra single frame in 8 * N + 1

Issue - State: closed - Opened by zhang-haojie about 2 months ago - 4 comments

#637 - CogVideoX-1.5-5B Inference all black when fp16

Issue - State: closed - Opened by cococozibojia about 2 months ago - 1 comment

#636 - diffusion_pytorch_model.bin missing

Issue - State: open - Opened by MrBlindshot about 2 months ago - 2 comments

#635 - cogvideox 5b full fine-tune OOM

Issue - State: open - Opened by linwenzhao1 about 2 months ago - 1 comment

#635 - cogvideox 5b full fine-tune OOM

Issue - State: open - Opened by linwenzhao1 about 2 months ago

#634 - How can we use LoRAs with current pipeline?

Issue - State: closed - Opened by FurkanGozukara about 2 months ago - 3 comments

#633 - cogvideo 2b full fine-tune model conversion

Issue - State: open - Opened by linwenzhao1 about 2 months ago - 1 comment

#633 - cogvideo 2b full fine-tune model conversion

Issue - State: open - Opened by linwenzhao1 about 2 months ago - 1 comment

#632 - Refactored the training code of finetune

Pull Request - State: closed - Opened by OleehyO about 2 months ago

#630 - No change for total loss during finetuning using SAT

Issue - State: open - Opened by qidai2000 about 2 months ago - 1 comment

#628 - CogVideoX-5B-I2V and CogVideoX1.5 diffusers finetune

Issue - State: open - Opened by JonathanLi19 about 2 months ago - 1 comment

#627 - Dataloader 没有真正的 shuffle

Issue - State: open - Opened by chunyu-li about 2 months ago

#626 - CogVideoX-5B-T2V black screen running colab

Issue - State: open - Opened by kazakovaanastasia about 2 months ago - 3 comments

#625 - Model conversion error

Issue - State: closed - Opened by linwenzhao1 about 2 months ago - 1 comment

#622 - cogvideox 1.5 lora train

Issue - State: closed - Opened by Passenger12138 about 2 months ago
Labels: good first issue

#621 - 咨询cogvideox1.5训练和推理框架支持现状

Issue - State: open - Opened by wmm-coding about 2 months ago - 5 comments

#620 - Not working on ComfyUI anymore

Issue - State: closed - Opened by icemagno 2 months ago - 1 comment

#618 - cogvideox微调的代码选择

Issue - State: closed - Opened by AmazDeng 2 months ago - 4 comments

#617 - Can't use xformers in "THUDM/CogVideoX1.5-5B-I2V"

Issue - State: closed - Opened by Eom-taeseon 2 months ago - 1 comment

#616 - Human Evaluation Ambiguity

Issue - State: open - Opened by antragoudaras 2 months ago - 2 comments

#615 - Cog video x dev

Pull Request - State: closed - Opened by OleehyO 2 months ago

#614 - 咨询下1.5的帧数问题

Issue - State: closed - Opened by wmm-coding 2 months ago - 10 comments
Labels: sat

#613 - Missing pipe on samples, can't get to the decoder

Issue - State: open - Opened by AinzAkaOver 2 months ago - 2 comments

#609 - Question about the context parallel setting in Video VAE

Issue - State: closed - Opened by Chester-CS 2 months ago - 4 comments

#608 - Cars always move backward

Issue - State: closed - Opened by jessemaxh 2 months ago

#607 - Add resolution warning

Pull Request - State: closed - Opened by OleehyO 2 months ago

#606 - run interpolation based on cogVideoX1.5B (i2v model convert)

Issue - State: closed - Opened by Suprhimp 2 months ago - 1 comment

#605 - The CogVideoX1.5-5B-I2V of diffusers

Issue - State: closed - Opened by liuxiaoyu1104 2 months ago