Mobius Labs GmbH

Team

company

https://www.mobiuslabs.com/

mobiusml

Activity Feed

AI & ML interests

Computer Vision, LLMs, Multimodal Models, Model Compression

Recent Activity

mobicham new activity about 6 hours ago

mobiuslabsgmbh/Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit:🙋🏻‍♂️ gemlite hqq quant version "unlearns" tool calling (?)

mobicham new activity about 7 hours ago

mobiuslabsgmbh/Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit:🙋🏻‍♂️ gemlite hqq quant version "unlearns" tool calling (?)

View all activity

mobicham

in mobiuslabsgmbh/Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit about 6 hours ago

🙋🏻‍♂️ gemlite hqq quant version "unlearns" tool calling (?)

#2 opened about 7 hours ago by

Tonic

appoose

updated a model 4 days ago

mobiuslabsgmbh/Qwen2.5-VL-7B-Instruct-leftpad

Updated 4 days ago

appoose

published a model 4 days ago

mobiuslabsgmbh/Qwen2.5-VL-7B-Instruct-leftpad

Updated 4 days ago

appoose

posted an update 12 months ago

Post

2102

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 12 months ago

Post

1798

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