馃憢馃徑 Welcome and Thank you for taking the time to check out TITAN
I know what went through your mind ... Yes.. It's another merge.. but I think you will really enjoy the experience.
I put alot of love and thought into how this was merged. Enjoy friend. Please give me feed back and I can't wait to see what you create 馃帹
-- Here is a GPT I created to help you with Mech Architecting :
https://chat.openai.com/g/g-tAGrWzxtt-titan-mech-architect
I'm seeing a trend of creators not showing the merged models at all and to be honest it doesn't really bother me. I don't know if this is a shared practice we are moving forward with but I understand the work and time that goes into a model .. trained or merged.. SO 馃憞
Please check out the other creators from the models I love and chose for my SDXL base model master merge:
Ingredients:
Bmegamix
Cheyenne
Cinevision
Colorful 2.6
CreaturesV2
Cyberrealistic
DuchaitenAiArt
Dynavision AIO
Everly Heights
Faetasticv 2.4
Inside Job Cartoon Mix
Juggernaut v9 + RunDiffusionPhoto 2
Mecha-XL
Modern Disney V3
Poltergeist
Polyhedron 3
Protovision 3D
Ratatoskr (Animal, Creature and Furry)
RelicC
SDXLyamersPerfectUnlimited
Tanukikanawa Classic Comic
Wildcard Animation V2
xVision v7
ZBase-XL - High Quality Aesthetic Art and Photo
ZavyChromaV5
May I Suggest checking these LoRAs out, I really like them:LORA: Blade_Runner_2049_Loha (Im just now exploring it and it works well with 'TITAN'
LORA: Textures Pack - Over 100 Textures!
Description
Optional Trigger: MECHA, MODISN HYPERION RENDER
vae should be baked in, if not, it is the SDXL_VAE
FAQ
Comments (11)
I very much appreciate you listing out the models you've used - every merger should do that! It's best to give recognition to others who have put in the time to move the craft forward.
And you picked great models to merge, so I'm going to give this a go.
cannot load, lack several unet blocks, anything wrong?
I just finshed downloading the file and importing it. I can confirm it is working as it's supposed to. Im using Draw Things on Mac, No custom encoder and same vae
@Chadxai聽so it's possible to have some issue on forge webui or so...
Thanks for providing info
Hi, I agree with the previous commenter. This is not working for me in Comfy. Here is the error I receive when it tries to load the Titan checkpoint:
Error occurred when executing CheckpointLoader|pysssss: 'NoneType' object has no attribute 'supported_inference_dtypes' File "D:\ComfyUI\execution.py", line 151, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\execution.py", line 81, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\execution.py", line 74, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\custom_nodes\ComfyUI-Custom-Scripts\py\better_combos.py", line 124, in load_checkpoint return super().load_checkpoint(**kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\nodes.py", line 541, in load_checkpoint out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings")) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI\comfy\sd.py", line 508, in load_checkpoint_guess_config unet_dtype = model_management.unet_dtype(model_params=parameters, supported_dtypes=model_config.supported_inference_dtypes) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Same. Problems loading block, runtime errors....not working.
same as faux here with comfy.
I was told it has something to do with comfy calling on slightly different info. I don't have a clue how to "fix" it for confy users but i can suggest waiting for an update. Im using Draw Things Local Mac app. @Sarimae聽@darkcee聽
@SLAPaper I have not used Forge but I love your desire to try the model. Im not familiar with the differences of the files and how each ui handles them. I'm currently working with Draw Things in my workflow
聽
Just so you know it isn't a Comfy thing.... I can't use it in Automatic 1111 1.7 on the PC either. And I was a diehard Agartha user, so I really wanted this to work.
RuntimeError: Error(s) in loading state_dict for LatentDiffusion: size mismatch for model.diffusion_model.input_blocks.4.1.transformer_blocks.0.attn2.to_k.weight: copying a param with shape torch.Size([640, 2048]) from checkpoint, the shape in current model is torch.Size([640, 768]).
I can send you the full error list if it would help. I have over 140 different XL models and this is the only one that doesn't work.
yes please send it to me
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.