AI & ML interests
None defined yet.
Recent Activity
Papers
Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-Experts
Agentic Confidence Calibration
xLAM: A Family of Large Action Models to Empower AI Agent Systems: https://github.com/SalesforceAIResearch/xLAM
FARE are Salesforce AI Research's open multi-task evaluator models.
A collection of GUI grounding models trained with GRPO.
This collection contains the best-performing TACO models based on LLaMA-3/Qwen2 and SigLIP/CLIP.
A collection of all BLIP models
-
Salesforce/blip-image-captioning-large
Image-to-Text • 0.5B • Updated • 719k • 1.45k -
Salesforce/blip-image-captioning-base
Image-to-Text • Updated • 2.43M • 842 -
Salesforce/blip-vqa-base
Visual Question Answering • 0.4B • Updated • 654k • 187 -
Salesforce/blip-vqa-capfilt-large
Visual Question Answering • Updated • 25.8k • 53
A collection that contains all InstructBLIP models!
-
Salesforce/instructblip-vicuna-7b
Image-Text-to-Text • 8B • Updated • 14.2k • 99 -
Salesforce/instructblip-vicuna-13b
Image-Text-to-Text • 14B • Updated • 553 • 43 -
Salesforce/instructblip-flan-t5-xxl
Image-Text-to-Text • 12B • Updated • 560 • 21 -
Salesforce/instructblip-flan-t5-xl
Image-Text-to-Text • Updated • 10.7k • 30
A collection of embedding models
-
Salesforce/SFR-Embedding-2_R
Feature Extraction • 7B • Updated • 280k • 94 -
Salesforce/SFR-Embedding-Mistral
Feature Extraction • 7B • Updated • 8.74k • 296 -
Salesforce/SFR-Embedding-Code-2B_R
Feature Extraction • Updated • 42.1k • 48 -
Salesforce/SFR-Embedding-Code-400M_R
Feature Extraction • 0.4B • Updated • 19.4k • 34
A collection's of Salesforce's Finance-specific model
CoDA is Salesforce AI Research's open, lightweight and diffusion-based language model.
SweRank is a framework for software issue localization, combining an embedding-based retriever (SweRankEmbed) with an LLM-based reranker (SweRankLLM).
A collection of all XGen-MM (Foundation LMM) models!
-
Salesforce/xgen-mm-phi3-mini-instruct-interleave-r-v1.5
Image-Text-to-Text • Updated • 799 • 59 -
Salesforce/blip3-kale
Viewer • Updated • 235M • 1.83k • 45 -
Salesforce/xgen-mm-vid-phi3-mini-r-v1.5-128tokens-8frames
Image-Text-to-Text • 4B • Updated • 29 • 11 -
Salesforce/xgen-mm-phi3-mini-instruct-r-v1
Image-Text-to-Text • 5B • Updated • 658 • 186
This collection contains all versions of the CoTA (Chain-of-Thought-and-Action) datasets.
A collection of contextual QA datasets dedicated to evaluate the contextual faithfulness of LLMs
A collection of all BLIP2 models!
-
Salesforce/blip2-opt-2.7b
Image-Text-to-Text • 4B • Updated • 460k • 432 -
Salesforce/blip2-flan-t5-xxl
Image-Text-to-Text • 12B • Updated • 3.58k • 93 -
Salesforce/blip2-opt-6.7b-coco
Image-Text-to-Text • 8B • Updated • 719 • 34 -
Salesforce/blip2-opt-6.7b
Image-Text-to-Text • 8B • Updated • 25k • 80
-
Salesforce/moirai-2.0-R-small
Time Series Forecasting • Updated • 216k • 37 -
Salesforce/moirai-moe-1.0-R-base
Time Series Forecasting • Updated • 174k • 16 -
Salesforce/moirai-1.1-R-small
Time Series Forecasting • Updated • 44.1k • 7 -
Salesforce/moirai-1.1-R-base
Time Series Forecasting • Updated • 23.7k • 7
A collection's of Salesforce's Finance-specific model
xLAM: A Family of Large Action Models to Empower AI Agent Systems: https://github.com/SalesforceAIResearch/xLAM
FARE are Salesforce AI Research's open multi-task evaluator models.
CoDA is Salesforce AI Research's open, lightweight and diffusion-based language model.
A collection of GUI grounding models trained with GRPO.
SweRank is a framework for software issue localization, combining an embedding-based retriever (SweRankEmbed) with an LLM-based reranker (SweRankLLM).
A collection of all XGen-MM (Foundation LMM) models!
-
Salesforce/xgen-mm-phi3-mini-instruct-interleave-r-v1.5
Image-Text-to-Text • Updated • 799 • 59 -
Salesforce/blip3-kale
Viewer • Updated • 235M • 1.83k • 45 -
Salesforce/xgen-mm-vid-phi3-mini-r-v1.5-128tokens-8frames
Image-Text-to-Text • 4B • Updated • 29 • 11 -
Salesforce/xgen-mm-phi3-mini-instruct-r-v1
Image-Text-to-Text • 5B • Updated • 658 • 186
This collection contains all versions of the CoTA (Chain-of-Thought-and-Action) datasets.
This collection contains the best-performing TACO models based on LLaMA-3/Qwen2 and SigLIP/CLIP.
A collection of contextual QA datasets dedicated to evaluate the contextual faithfulness of LLMs
A collection of all BLIP models
-
Salesforce/blip-image-captioning-large
Image-to-Text • 0.5B • Updated • 719k • 1.45k -
Salesforce/blip-image-captioning-base
Image-to-Text • Updated • 2.43M • 842 -
Salesforce/blip-vqa-base
Visual Question Answering • 0.4B • Updated • 654k • 187 -
Salesforce/blip-vqa-capfilt-large
Visual Question Answering • Updated • 25.8k • 53
A collection of all BLIP2 models!
-
Salesforce/blip2-opt-2.7b
Image-Text-to-Text • 4B • Updated • 460k • 432 -
Salesforce/blip2-flan-t5-xxl
Image-Text-to-Text • 12B • Updated • 3.58k • 93 -
Salesforce/blip2-opt-6.7b-coco
Image-Text-to-Text • 8B • Updated • 719 • 34 -
Salesforce/blip2-opt-6.7b
Image-Text-to-Text • 8B • Updated • 25k • 80
A collection that contains all InstructBLIP models!
-
Salesforce/instructblip-vicuna-7b
Image-Text-to-Text • 8B • Updated • 14.2k • 99 -
Salesforce/instructblip-vicuna-13b
Image-Text-to-Text • 14B • Updated • 553 • 43 -
Salesforce/instructblip-flan-t5-xxl
Image-Text-to-Text • 12B • Updated • 560 • 21 -
Salesforce/instructblip-flan-t5-xl
Image-Text-to-Text • Updated • 10.7k • 30
-
Salesforce/moirai-2.0-R-small
Time Series Forecasting • Updated • 216k • 37 -
Salesforce/moirai-moe-1.0-R-base
Time Series Forecasting • Updated • 174k • 16 -
Salesforce/moirai-1.1-R-small
Time Series Forecasting • Updated • 44.1k • 7 -
Salesforce/moirai-1.1-R-base
Time Series Forecasting • Updated • 23.7k • 7
A collection of embedding models
-
Salesforce/SFR-Embedding-2_R
Feature Extraction • 7B • Updated • 280k • 94 -
Salesforce/SFR-Embedding-Mistral
Feature Extraction • 7B • Updated • 8.74k • 296 -
Salesforce/SFR-Embedding-Code-2B_R
Feature Extraction • Updated • 42.1k • 48 -
Salesforce/SFR-Embedding-Code-400M_R
Feature Extraction • 0.4B • Updated • 19.4k • 34