MXFP4/NVFP4 models
AI & ML interests
Computer Vision, LLMs, Multimodal Models, Model Compression
HQQ-quantized Qwen models
-
dropbox-dash/Qwen2.5-7B-Instruct-1M_4bitgs64_hqq_hf
Text Generation • Updated • 9 • 1 -
dropbox-dash/Qwen2.5-VL-7B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 6 • 1 -
dropbox-dash/Qwen2.5-VL-3B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 11 • 1 -
dropbox-dash/Qwen2.5-7B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 13 • 2
-
dropbox-dash/Llama-3.1-8b-instruct_4bitgs64_hqq_calib
Text Generation • Updated • 8 • 55 -
dropbox-dash/Llama-3.1-8B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 14 • 1 -
dropbox-dash/lama-3.1-70b-instruct_4bitgs64_hqq
Text Generation • Updated • 18 • 32 -
dropbox-dash/Llama-3.2-3B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 12 • 1
LLama2 models quantized using https://github.com/mobiusml/hqq
-
mobiuslabsgmbh/Llama-2-7b-hf-4bit_g64-HQQ
Text Generation • Updated • 38 • 2 -
mobiuslabsgmbh/Llama-2-13b-hf-4bit_g64-HQQ
Text Generation • Updated • 28 • 1 -
mobiuslabsgmbh/Llama-2-70b-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 50 • 2 -
mobiuslabsgmbh/Llama-2-70b-chat-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 44 • 3
ViT models quantized using https://github.com/mobiusml/hqq
Quantized models in AO/GemLite format
-
dropbox-dash/Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 5 • 2 -
dropbox-dash/Phi-4-mini-instruct_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 8 • 1 -
dropbox-dash/Qwen2.5-7B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 10 • 2 -
dropbox-dash/Qwen3-32B_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 10 • 1
Re-distilled DeepSeek R1 models
-
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Llama3-8B-v1.1
Text Generation • 8B • Updated • 62 • • 13 -
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-7B-v1.1
Text Generation • 8B • Updated • 60 • 16 -
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.1
Text Generation • 2B • Updated • 102 • 13 -
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.0
Text Generation • 2B • Updated • 83 • 44
This collection will include language, vision and audio models pre-trained or fine-tuned by Mobius Labs GmbH
4-bit and 2-bit Mixtral models quantized using https://github.com/mobiusml/hqq
-
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-4bit_g64-HQQ
Text Generation • Updated • 30 • 9 -
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 61 • 9 -
mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 52 • 4 -
mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-4bit_g64-HQQ
Text Generation • Updated • 28 • 1
MXFP4/NVFP4 models
Quantized models in AO/GemLite format
-
dropbox-dash/Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 5 • 2 -
dropbox-dash/Phi-4-mini-instruct_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 8 • 1 -
dropbox-dash/Qwen2.5-7B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 10 • 2 -
dropbox-dash/Qwen3-32B_gemlite-ao_a16w4_gs_128_pack_32bit
Text Generation • Updated • 10 • 1
HQQ-quantized Qwen models
-
dropbox-dash/Qwen2.5-7B-Instruct-1M_4bitgs64_hqq_hf
Text Generation • Updated • 9 • 1 -
dropbox-dash/Qwen2.5-VL-7B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 6 • 1 -
dropbox-dash/Qwen2.5-VL-3B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 11 • 1 -
dropbox-dash/Qwen2.5-7B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 13 • 2
Re-distilled DeepSeek R1 models
-
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Llama3-8B-v1.1
Text Generation • 8B • Updated • 62 • • 13 -
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-7B-v1.1
Text Generation • 8B • Updated • 60 • 16 -
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.1
Text Generation • 2B • Updated • 102 • 13 -
mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.0
Text Generation • 2B • Updated • 83 • 44
-
dropbox-dash/Llama-3.1-8b-instruct_4bitgs64_hqq_calib
Text Generation • Updated • 8 • 55 -
dropbox-dash/Llama-3.1-8B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 14 • 1 -
dropbox-dash/lama-3.1-70b-instruct_4bitgs64_hqq
Text Generation • Updated • 18 • 32 -
dropbox-dash/Llama-3.2-3B-Instruct_4bitgs64_hqq_hf
Text Generation • Updated • 12 • 1
This collection will include language, vision and audio models pre-trained or fine-tuned by Mobius Labs GmbH
LLama2 models quantized using https://github.com/mobiusml/hqq
-
mobiuslabsgmbh/Llama-2-7b-hf-4bit_g64-HQQ
Text Generation • Updated • 38 • 2 -
mobiuslabsgmbh/Llama-2-13b-hf-4bit_g64-HQQ
Text Generation • Updated • 28 • 1 -
mobiuslabsgmbh/Llama-2-70b-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 50 • 2 -
mobiuslabsgmbh/Llama-2-70b-chat-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 44 • 3
4-bit and 2-bit Mixtral models quantized using https://github.com/mobiusml/hqq
-
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-4bit_g64-HQQ
Text Generation • Updated • 30 • 9 -
mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 61 • 9 -
mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-2bit_g16_s128-HQQ
Text Generation • Updated • 52 • 4 -
mobiuslabsgmbh/Mixtral-8x7B-v0.1-hf-4bit_g64-HQQ
Text Generation • Updated • 28 • 1
ViT models quantized using https://github.com/mobiusml/hqq