arcee-ai mergekit issues

arcee-ai / mergekit

Tools for merging pretrained large language models.

GNU Lesser General Public License v3.0

4.88k stars 446 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Pad embeds to multiple

#465 cg123 closed 1 day ago
0
Better tied weight handling

#464 cg123 closed 1 day ago
0
Handle optional weights in mergekit-moe

#463 cg123 closed 1 day ago
0
Rewrite readme more novice-friendly

#462 clover1980 opened 4 days ago
0
mergekit for vision models

#461 prince0310 opened 5 days ago
0
Why are the names of parameters hard-coded? Is it possible to read it from index.json in HF checkpoints?

#460 zhangzx-uiuc opened 1 week ago
1
Qwen2.5 LoRA Extraction not working in vLLM & Aphrodite Engine

#459 Nero10578 opened 1 week ago
2
add whisper model

#458 sagewe opened 2 weeks ago
0
Critical Merging Bug just started...

#457 David-AU-github opened 2 weeks ago
1
About Model-Breadcrumbs merge implementation

#455 vishaal27 opened 3 weeks ago
0
Base Model generation time increases when passed through the MergeKit

#454 ahmedamrelhefnawy opened 3 weeks ago
0
N-model ModelStock merging

#453 vishaal27 opened 3 weeks ago
1
Moe merging failed

#452 PsoriasiIR opened 3 weeks ago
2
Use sst2 to eval merging

#451 VivianeGalvao closed 4 weeks ago
0
Merge Models with Non-Standard Architectures (e.g., Multimodal Models)

#450 ElliotStein opened 1 month ago
3
[question] multi gpu available?

#449 eunbin079 opened 1 month ago
0
Bump version number

#448 cg123 closed 1 month ago
0
mergekit-extract-lora does not extract - the destination is empty

#447 raulod opened 1 month ago
2
KeyError model[0] did not exist in tensor?

#446 FrozzDay opened 1 month ago
2
Report issues regarding the architecture-agnostic branch.

#445 win10ogod opened 1 month ago
3
Bump dependencies

#444 cg123 closed 1 month ago
0
RuntimeError: Need to specify cache dir to merge adapters

#442 Zolilio closed 1 month ago
1
Add methods from https://arxiv.org/abs/2405.07813

#441 zsgvivo closed 1 day ago
2
add methods from https://arxiv.org/abs/2405.07813

#440 zsgvivo closed 1 month ago
0
11

#439 meiyiyeshi closed 1 month ago
0
[question] `task_arithmetic` simple question

#438 eunbin079 closed 1 month ago
2
After the two Qwen1.5-7B-chat models were merged, garbled inference results appeared.

#437 Zhangfanfan0101 closed 1 month ago
0
Fixed the YML/YAML documentation for Qwen MoE creation

#435 Nottlespike opened 1 month ago
1
[request] Support for Vision Language Models

#434 NickGao96 closed 1 month ago
13
[request]Can it support architectures such as stable diffusion Xl and flux dev?

#433 win10ogod opened 1 month ago
2
Initial implementation of PCB merge method

#432 cg123 opened 1 month ago
0
Update actors.py

#431 kwon13 opened 1 month ago
0
Handle merges stored as list instead of space-separated string

#430 cg123 closed 1 month ago
0
Update Llama architecture to handle 3b/1b

#429 cg123 closed 1 month ago
0
Broken tokenizer in Yi-34B merge

#428 Asherathe closed 1 month ago
3
I would like to merge the deepseekForCausalLM model. Are there any related examples available?

#427 xaiocaibi opened 2 months ago
0
Merging Lora fine-tuned models with MoE

#426 AmineBechar07 opened 2 months ago
0
Qwen2.5 14B models are ... sometimes? ... having their token vocabulary truncated down to 'actual'?

#425 ann-brown opened 2 months ago
4
Support for new Llama 3.2 - 1B / 3B ?

#424 David-AU-github closed 1 month ago
13
Support for Vision Model such as ViT

#423 redagavin opened 2 months ago
0
Support for xlm-roberta

#422 umiron opened 2 months ago
2
"mergekit-yaml" not created upon installation

#421 BovineOverlord opened 2 months ago
2
How to use multi GPUs

#420 liudan193 opened 2 months ago
1
would you like to support Qwen2.5 Model?

#419 ArcherShirou closed 2 months ago
1
Input should be a valid dictionary or instance of MergeConfiguration

#418 Hugo-Calero opened 2 months ago
2
Make Cohere lm_head optional

#417 cg123 closed 2 months ago
0
Add Solar And Exaone Model

#416 shing100 closed 2 months ago
1
Add support Exaone Model

#415 shing100 closed 2 months ago
2
Re-Train every block with reduced width

#414 snapo closed 2 months ago
0
Fix README links

#413 cg123 closed 3 months ago
0