Mobius Labs GmbH

company

https://www.mobiuslabs.com/

mobiusml

Activity Feed

AI & ML interests

Computer Vision, LLMs, Multimodal Models, Model Compression

Recent Activity

Jilt updated a model 3 months ago

mobiuslabsgmbh/faster-whisper-large-v3-turbo

mobicham updated a model 4 months ago

mobiuslabsgmbh/Llama-2-7b-chat-hf_1bitgs8_hqq

mobicham updated a model 5 months ago

mobiuslabsgmbh/Llama-3.1-8b-instruct_4bitgs64_hqq

View all activity

mobiuslabsgmbh's activity

Jilt

updated a model 3 months ago

mobiuslabsgmbh/faster-whisper-large-v3-turbo

Updated Oct 8, 2024 • 39.2k • 16

mobicham

updated a model 4 months ago

mobiuslabsgmbh/Llama-2-7b-chat-hf_1bitgs8_hqq

Text Generation • Updated Sep 10, 2024 • 97 • 75

mobicham

updated 2 models 5 months ago

mobiuslabsgmbh/Llama-3.1-8b-instruct_4bitgs64_hqq

Text Generation • Updated Aug 27, 2024 • 18 • 1

mobiuslabsgmbh/Llama-3.1-8b-instruct_4bitgs64_hqq_calib

Text Generation • Updated Aug 27, 2024 • 67 • 55

appoose

updated a model 5 months ago

mobiuslabsgmbh/Hermes-3-Llama-3.1-70B_4bitgs64_hqq

Text Generation • Updated Aug 16, 2024 • 4 • 4

mobicham

updated 3 models 5 months ago

appoose

posted an update 5 months ago

Post

2067

Releasing HQQ Llama-3.1-70b 4-bit quantized version! Check it out at mobiuslabsgmbh/Llama-3.1-70b-instruct_4bitgs64_hqq.

Achieves 99% of the base model performance across various benchmarks! Details in the model card.

appoose

posted an update 6 months ago

Post

1790

Excited to announce the release of our high-quality Llama-3.1 8B 4-bit HQQ/calibrated quantized model! Achieving an impressive 99.3% relative performance to FP16, it also delivers the fastest inference speed for transformers.

mobiuslabsgmbh/Llama-3.1-8b-instruct_4bitgs64_hqq_calib