LLM EXPLORER 60,587 MODELS INDEXED

Zephyr 7B Ipo 0K 15K I1 by DUAL-GPO

By DUAL-GPO · 5 downloads
HF Repo ↗ Copy URL

Zephyr 7B Ipo 0K 15K I1 is an open-source language model by DUAL-GPO. Features: 7b LLM, VRAM: 0.7GB, LLM Explorer Score: 0.18.

  Adapter   Alignment-handbook Base model:adapter:dual-gpo/ze... Base model:dual-gpo/zephyr-7b-... Dataset:huggingfaceh4/ultrafee...   Dpo   Finetuned   Generated from trainer   Lora   Mistral   Peft   Region:us   Safetensors   Tensorboard   Trl
Model Card on HF 🤗: https://huggingface.co/DUAL-GPO/zephyr-7b-ipo-0k-15k-i1 

Zephyr 7B Ipo 0K 15K I1 Parameters and Internals

LLM NameZephyr 7B Ipo 0K 15K I1
Repository 🤗https://huggingface.co/DUAL-GPO/zephyr-7b-ipo-0k-15k-i1 
Base Model(s)  Zephyr 7B Ipo Qlora V0 Merged   DUAL-GPO/zephyr-7b-ipo-qlora-v0-merged
Model Size7b
Required VRAM0.7 GB
Updated2026-09-07
MaintainerDUAL-GPO
Model Files  0.7 GB   0.0 GB
Model ArchitectureAdapter
Model Max Length2048
Is Biasednone
Tokenizer ClassLlamaTokenizer
Padding Token</s>
PEFT TypeLORA
LoRA ModelYes
PEFT Target Moduleso_proj|gate_proj|up_proj|k_proj|v_proj|down_proj|q_proj
LoRA Alpha128
LoRA Dropout0.05
R Param128

Best Alternatives to Zephyr 7B Ipo 0K 15K I1

Best Alternatives
Context / RAM
Downloads
Likes
K2 Horizon 7B Uno0K / 1.4 GB1010516
...x H3 Prompt Rewriter LoRA Omni0K / 1.3 GB2732
DeepMath R1 Distill Qwen 7B0K / 0.1 GB131
Llama2 7B Qlora Guanaco0K / 0.1 GB111
Qwen Megumin0K / 0.1 GB111
Mistral 7B Instruct Vi Alpaca0K / 0.4 GB50
...s 25 Mistral 7B Irca DPO Pairs0K / 0.1 GB60
Qwen1.5 7B Chat Sa V0.10K / 0 GB80
Uk Fraud Chatbot Llama20K / 0.4 GB60
Deepseek Llm 7B Chat Sa V0.10K / 0 GB70
Note: green Score (e.g. "73.2") means that the model is better than DUAL-GPO/zephyr-7b-ipo-0k-15k-i1.