Introducing AutoRound: Intel’s Advanced Quantization for LLMs and VLMs
•
39
Intel
company
Verified
AI & ML interests
None defined yet.
Recent Activity
View all activity
Articles
Text generation LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Audio Classification models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
NLP models for Feature Extraction that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
BAAI/bge-base-en-v1.5
Feature Extraction • 0.1B • Updated • 3.03M • • 334 -
BAAI/bge-large-en-v1.5
Feature Extraction • 0.3B • Updated • 3.49M • • 560 -
Contrastive-Tension/BERT-Large-CT-STSb
Feature Extraction • Updated • 32 -
DeepPavlov/bert-base-cased-conversational
Feature Extraction • Updated • 338 • • 8
Image-to-text models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
LLMs for Question Answering that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
aware-ai/roberta-large-squadv2
Question Answering • Updated • 4 -
deepset/bert-base-cased-squad2
Question Answering • 0.1B • Updated • 60k • 20 -
deepset/roberta-base-squad2
Question Answering • 0.1B • Updated • 1.16M • • 905 -
distilbert/distilbert-base-uncased-distilled-squad
Question Answering • 0.1B • Updated • 127k • • 116
Text2Text Generation LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
LLMs for translation tasks that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Fine-tuned 7B parameter LLM models, one of which made it to the top of the 7B HF LLM Leaderboard
DPT 3.0 (MiDaS) models, leveraging ViT and ViT-hybrid backbones
Text-Visual Prompting
Suite of diffusion models targeting virtual reality development
Smaller BERT models for question answering and text classification
-
Intel/distilbert-base-cased-distilled-squad-int8-static-inc
Question Answering • Updated • 8 -
Intel/distilbert-base-uncased-MRPC-int8-dynamic-inc
Text Classification • Updated • 4 • 1 -
Intel/distilbert-base-uncased-MRPC-int8-static-inc
Text Classification • Updated • 5 -
Intel/distilbert-base-uncased-distilled-squad-int8-static-inc
Question Answering • Updated • 1.57k • 5
DeBERTa is a language model that originates from Meta's RoBERTa model with disentangled attention and enhanced mask decoder.
Text retrieval model, trained on the Natural Questions dataset
Fine-tuned version of Microsoft's MiniLM models, trained on the GLUE MRPC dataset.
-
Intel/MiniLM-L12-H384-uncased-mrpc
Text Classification • Updated • 17 • 1 -
Intel/MiniLM-L12-H384-uncased-mrpc-int8-dynamic-inc
Text Classification • Updated • 16 -
Intel/MiniLM-L12-H384-uncased-mrpc-int8-qat-inc
Text Classification • Updated • 16 -
Intel/MiniLM-L12-H384-uncased-mrpc-int8-static-inc
Text Classification • Updated • 16
Large scale information retrieval corpus that was created based on real user search queries using Bing search engine
Originally from Google: Text-To-Text Transfer Transformer (T5)
Original paper: XLNet: Generalized Autoregressive Pretraining for Language Understanding
This collection contains the models, papers, and demo associated with the LDM3D release.
Text-to-image models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Automatic Speech Recognition models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
openai/whisper-small
Automatic Speech Recognition • 0.2B • Updated • 1.98M • 436 -
distil-whisper/distil-medium.en
Automatic Speech Recognition • 0.4B • Updated • 36.9k • 124 -
facebook/hubert-large-ls960-ft
Automatic Speech Recognition • Updated • 347k • 74 -
openai/whisper-base
Automatic Speech Recognition • 0.1B • Updated • 535k • 237
Image Classification models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Masked language models (MLMs) that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Text Classification LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
Alireza1044/albert-base-v2-sst2
Text Classification • Updated • 35 -
BAAI/bge-reranker-base
Text Classification • 0.3B • Updated • 1.22M • 205 -
ChrisZeng/electra-large-discriminator-nli-efl-tweeteval
Text Classification • Updated • 15 -
DeepPavlov/xlm-roberta-large-en-ru-mnli
Text Classification • Updated • 43 • 2
Token Classification LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
FacebookAI/xlm-roberta-large-finetuned-conll03-english
Token Classification • 0.6B • Updated • 110k • • 172 -
Jean-Baptiste/roberta-large-ner-english
Token Classification • 0.4B • Updated • 97.4k • • 74 -
dslim/bert-base-NER
Token Classification • 0.1B • Updated • 1.48M • • 634 -
dslim/bert-large-NER
Token Classification • 0.3B • Updated • 83.1k • • 155
DPT 3.1 (MiDaS) models, leveraging state-of-the-art vision backbones such as BEiT and Swinv2
-
MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation
Paper • 2307.14460 • Published • 9 -
Intel/dpt-beit-large-512
Depth Estimation • 0.3B • Updated • 1.59k • 8 -
Intel/dpt-beit-large-384
Depth Estimation • 0.3B • Updated • 95 -
Intel/dpt-beit-base-384
Depth Estimation • 0.1B • Updated • 5.35k • 1
Whisper models for automatic speech recognition (ASR) and speech translation, quantized for faster inference speeds.
-
Intel/whisper-base-int8-dynamic-inc
Automatic Speech Recognition • Updated • 12 • 1 -
Intel/whisper-base-int8-static-inc
Automatic Speech Recognition • Updated • 12 -
Intel/whisper-base-onnx-int4-inc
Automatic Speech Recognition • Updated • 13 • 9 -
Intel/whisper-large-int8-dynamic-inc
Automatic Speech Recognition • Updated • 17 • 1
Series of GPT fine-tuned models
BERT models of varying flavors
-
Intel/bert-base-cased-finetuned-sst2-int8-inc
Text Classification • Updated • 14 -
Intel/bert-base-uncased-CoLA-int8-inc
Text Classification • Updated • 14 -
Intel/bert-base-uncased-QNLI-int8-inc
Text Classification • Updated • 15 -
Intel/bert-base-uncased-STS-B-int8-inc
Text Classification • Updated • 16
Quantized versions of ALBERT models for language tasks
Based on Metas's RoBERTa model released in 2019, trained on 138GB of French text.
Question Answering model, trained on the SQuAD 1.1 dataset
Adaptations on Meta's BART model
Natural Questions
Originally from Google, Vision Transformer (ViT)
Text generation LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Text-to-image models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Audio Classification models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Automatic Speech Recognition models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
openai/whisper-small
Automatic Speech Recognition • 0.2B • Updated • 1.98M • 436 -
distil-whisper/distil-medium.en
Automatic Speech Recognition • 0.4B • Updated • 36.9k • 124 -
facebook/hubert-large-ls960-ft
Automatic Speech Recognition • Updated • 347k • 74 -
openai/whisper-base
Automatic Speech Recognition • 0.1B • Updated • 535k • 237
NLP models for Feature Extraction that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
BAAI/bge-base-en-v1.5
Feature Extraction • 0.1B • Updated • 3.03M • • 334 -
BAAI/bge-large-en-v1.5
Feature Extraction • 0.3B • Updated • 3.49M • • 560 -
Contrastive-Tension/BERT-Large-CT-STSb
Feature Extraction • Updated • 32 -
DeepPavlov/bert-base-cased-conversational
Feature Extraction • Updated • 338 • • 8
Image Classification models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Image-to-text models that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Masked language models (MLMs) that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
LLMs for Question Answering that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
aware-ai/roberta-large-squadv2
Question Answering • Updated • 4 -
deepset/bert-base-cased-squad2
Question Answering • 0.1B • Updated • 60k • 20 -
deepset/roberta-base-squad2
Question Answering • 0.1B • Updated • 1.16M • • 905 -
distilbert/distilbert-base-uncased-distilled-squad
Question Answering • 0.1B • Updated • 127k • • 116
Text Classification LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
Alireza1044/albert-base-v2-sst2
Text Classification • Updated • 35 -
BAAI/bge-reranker-base
Text Classification • 0.3B • Updated • 1.22M • 205 -
ChrisZeng/electra-large-discriminator-nli-efl-tweeteval
Text Classification • Updated • 15 -
DeepPavlov/xlm-roberta-large-en-ru-mnli
Text Classification • Updated • 43 • 2
Text2Text Generation LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
Token Classification LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
-
FacebookAI/xlm-roberta-large-finetuned-conll03-english
Token Classification • 0.6B • Updated • 110k • • 172 -
Jean-Baptiste/roberta-large-ner-english
Token Classification • 0.4B • Updated • 97.4k • • 74 -
dslim/bert-base-NER
Token Classification • 0.1B • Updated • 1.48M • • 634 -
dslim/bert-large-NER
Token Classification • 0.3B • Updated • 83.1k • • 155
LLMs for translation tasks that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU.
DPT 3.1 (MiDaS) models, leveraging state-of-the-art vision backbones such as BEiT and Swinv2
-
MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimation
Paper • 2307.14460 • Published • 9 -
Intel/dpt-beit-large-512
Depth Estimation • 0.3B • Updated • 1.59k • 8 -
Intel/dpt-beit-large-384
Depth Estimation • 0.3B • Updated • 95 -
Intel/dpt-beit-base-384
Depth Estimation • 0.1B • Updated • 5.35k • 1
Fine-tuned 7B parameter LLM models, one of which made it to the top of the 7B HF LLM Leaderboard
Whisper models for automatic speech recognition (ASR) and speech translation, quantized for faster inference speeds.
-
Intel/whisper-base-int8-dynamic-inc
Automatic Speech Recognition • Updated • 12 • 1 -
Intel/whisper-base-int8-static-inc
Automatic Speech Recognition • Updated • 12 -
Intel/whisper-base-onnx-int4-inc
Automatic Speech Recognition • Updated • 13 • 9 -
Intel/whisper-large-int8-dynamic-inc
Automatic Speech Recognition • Updated • 17 • 1
Series of GPT fine-tuned models
DPT 3.0 (MiDaS) models, leveraging ViT and ViT-hybrid backbones
Text-Visual Prompting
Suite of diffusion models targeting virtual reality development
BERT models of varying flavors
-
Intel/bert-base-cased-finetuned-sst2-int8-inc
Text Classification • Updated • 14 -
Intel/bert-base-uncased-CoLA-int8-inc
Text Classification • Updated • 14 -
Intel/bert-base-uncased-QNLI-int8-inc
Text Classification • Updated • 15 -
Intel/bert-base-uncased-STS-B-int8-inc
Text Classification • Updated • 16
Smaller BERT models for question answering and text classification
-
Intel/distilbert-base-cased-distilled-squad-int8-static-inc
Question Answering • Updated • 8 -
Intel/distilbert-base-uncased-MRPC-int8-dynamic-inc
Text Classification • Updated • 4 • 1 -
Intel/distilbert-base-uncased-MRPC-int8-static-inc
Text Classification • Updated • 5 -
Intel/distilbert-base-uncased-distilled-squad-int8-static-inc
Question Answering • Updated • 1.57k • 5
Quantized versions of ALBERT models for language tasks
Based on Metas's RoBERTa model released in 2019, trained on 138GB of French text.
DeBERTa is a language model that originates from Meta's RoBERTa model with disentangled attention and enhanced mask decoder.
Text retrieval model, trained on the Natural Questions dataset
Question Answering model, trained on the SQuAD 1.1 dataset
Fine-tuned version of Microsoft's MiniLM models, trained on the GLUE MRPC dataset.
-
Intel/MiniLM-L12-H384-uncased-mrpc
Text Classification • Updated • 17 • 1 -
Intel/MiniLM-L12-H384-uncased-mrpc-int8-dynamic-inc
Text Classification • Updated • 16 -
Intel/MiniLM-L12-H384-uncased-mrpc-int8-qat-inc
Text Classification • Updated • 16 -
Intel/MiniLM-L12-H384-uncased-mrpc-int8-static-inc
Text Classification • Updated • 16
Adaptations on Meta's BART model
Natural Questions
Large scale information retrieval corpus that was created based on real user search queries using Bing search engine
Originally from Google: Text-To-Text Transfer Transformer (T5)
Original paper: XLNet: Generalized Autoregressive Pretraining for Language Understanding
Originally from Google, Vision Transformer (ViT)
This collection contains the models, papers, and demo associated with the LDM3D release.