Lightweight LLMs specialised in RAG based on t5, bart, bert, roberta with ONNX conversion
Tomás Marques
tomasmcm
AI & ML interests
Solutions Architect @ whitesmith.co ✦ Creative Technologist (Interaction Designer, Developer, Strategic Innovator)
Recent Activity
upvoted
a
collection
about 22 hours ago
GPT-OSS Pruned Experts (4.2B-20B) [IF, Science, Math, etc.]
new activity
about 22 hours ago
AmanPriyanshu/gpt-oss-11.4b-specialized-science-pruned-moe-only-16-experts:Coding variant
new activity
21 days ago
huihui-ai/DeepSeek-R1-Pruned-Coder-411B:Qwen/Qwen3-Coder-480B-A35B-Instruct Pruning
Organizations
None yet
models
25
tomasmcm/KwaiCoder-AutoThink-preview-mlx-3Bit
Text Generation
•
5B
•
Updated
•
21
tomasmcm/boltuix-neurobert-mini-onnx
Question Answering
•
Updated
•
5
tomasmcm/NeuroBERT-Mini-ONNX
Fill-Mask
•
Updated
•
3
tomasmcm/nlpconnect-deberta-v3-xsmall-squad2-onnx
Question Answering
•
Updated
•
6
tomasmcm/vmware-deberta-v3-base-mrqa-onnx
Question Answering
•
Updated
•
8
tomasmcm/vmware-tinyroberta-mrqa-onnx
Question Answering
•
Updated
•
4
tomasmcm/vmware-deberta-v3-xsmall-mrqa-onnx
Question Answering
•
Updated
•
6
tomasmcm/deepset-tinyroberta-squad2-onnx
Question Answering
•
Updated
•
5
tomasmcm/deepset-roberta-base-squad2-onnx
Question Answering
•
Updated
•
4
tomasmcm/intel-dynamic-tinybert-onnx
Question Answering
•
Updated
•
15
datasets
0
None public yet