Extreme low-bit quantization with HQQ+ (HQQ + LoRA adapter)
Mobius Labs GmbH
company
AI & ML interests
Computer Vision, LLMs, Multimodal Models, Model Compression
Organization Card
About org cards
Multimodal AI on a global scale. Advocates for Open Source and Open Intelligence. Currently investigating how to make Large Machine Learning Models smaller and democratize them for GPU-poor environments. Visit https://mobiusml.github.io/blog/ to see some of our recent work.
models
20
mobiuslabsgmbh/Llama-2-7b-chat-hf_4bitnogs_hqq
Text Generation
•
Updated
•
28
•
1
mobiuslabsgmbh/Llama-2-7b-chat-hf_2bitgs8_hqq
Text Generation
•
Updated
•
106
•
33
mobiuslabsgmbh/Llama-2-7b-chat-hf_1bitgs8_hqq
Text Generation
•
Updated
•
472
•
71
mobiuslabsgmbh/aanaphi2-v0.1
Text Generation
•
Updated
•
2.99k
•
22
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-attn-4bit-moe-3bit-metaoffload-HQQ
Text Generation
•
Updated
•
119
•
13
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-attn-4bit-moe-2bitgs8-metaoffload-HQQ
Text Generation
•
Updated
•
128
•
19
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-attn-4bit-moe-2bit-metaoffload-HQQ
Text Generation
•
Updated
•
93
•
14
mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-2bit_g16_s128-HQQ
Text Generation
•
Updated
•
9
•
4
mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-attn-4bit-moe-2bit-HQQ
Text Generation
•
Updated
•
9
•
6
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-2bit_g16_s128-HQQ
Text Generation
•
Updated
•
12
•
9