Downloads · 30 days
0
rag-models-mxl
LightSpeedtoOblivion/rag-models-mxl
rag-models-mxl is a machine learning model from LightSpeedtoOblivion. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
- e5distilledfp16ragclean - Text embeddings (224MB) - pleiasdistilledfp16ragclean - Text generation (190MB) - osmosisdistilledfp16clean - Advanced reasoning (2.2GB)
Access
Public
Updated Sep 13, 2025
Repo size
2.8 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors8.5 GB · 99%
At a glance
- Access
- Public
- Created
- Sep 13, 2025
- Updated
- Sep 13, 2025
- SHA
- c6324022
Model card
From the Hugging Face model README
RAG Models - MXL Quantized & FP16 Distilled
Models Included:
MXL Quantized Models (Apple Silicon Optimized):
- e5_distilled_fp16_rag_clean - Text embeddings (224MB)
- pleias_distilled_fp16_rag_clean - Text generation (190MB)
- osmosis_distilled_fp16_clean - Advanced reasoning (2.2GB)
Original FP16 Distilled Models:
- e5_distilled_fp16_rag - Original FP16 version
- pleias_distilled_fp16_rag - Original FP16 version
- osmosis_distilled_fp16 - Original FP16 version
Usage:
from mlx_lm import load
model, tokenizer = load('LightSpeedtoOblivion/rag-models-mxl/e5_distilled_fp16_rag_clean')
Architecture:
- E5: Sentence transformer for embeddings
- PLEIAS: LlamaForCausalLM for text generation
- OSMOSIS: Qwen3ForCausalLM for reasoning
S-DNS Naming:
- lse.colony.DoX.models.e5.mxl
- lse.colony.DoX.models.pleias.mxl
- lse.colony.DoX.models.osmosis.mxl
- Created
- Sep 13, 2025
- Updated
- Sep 13, 2025