LLMs

GTP-4 Turbo

GPT-4

Gemini

Lumiere by Google

GPT-3.5

Goody 2

OpenAI Text Moderation

Stabilityai - Expert ensemble, latent diffusion

Facebook - Repository English TTS model

NousResearch/Hermes-2-Pro-Mistral-7B-GGUF

Pyannote - Reproduce Interspeech 2021 results with segmentation

LLaMa 2

Distilbert - Distilled RoBERTa-base model

Dslim NER - Fine-tuned BERT, state-of-the-art NER

FacebookAI - English MLM pretrained model

PaLM 2

FacebookAI - English MLM pretrained model

upstage/SOLAR-10.7B-Instruct-v1.0

Nexusflow/Starling-LM-7B-beta

Facebook Opt - OPT introduced in May 2022

ReplaceAnything

Mistralai - Mistral-7B-Instruct-v0.2: Improved LLM

Distilbert - DistilBERT base, SST-2 fine-tuned

sentence t5-base - 768D vector space.

Weaver - Creative Writing Models

Google Byt5 - Tokenizer-free T5 version, architecture similar to MT5

Cross-encoder - Query encoding, passage matching

Pyannote speaker - Pipeline Sans onnxruntime, like pyannote

cerebras/Cerebras-GPT-111M

Openai-community - English CLM pretrained model

Stabilityai - Model card for Stable Diffusion v2

Marieke93 - Fine-tuned MiniLM on evidence types.

Pyannote - x-vector TDNN, SincNet features

Distiluse-base - Text to 512D vectors, clustering, semantic search

Openai Clip - Model trained from scratch

j-hartmann emotion - English text, Ekman's 6 emotions, trained on 6 datasets

Cardiffnlp - roBERTa-base, 58M tweets, TweetEval sentiment finetuned

anthropic/claude-3-haiku:beta

MoritzLaurer - Trained on MultiNLI, Fever-NLI, ANLI: 763,913 pairs

Contriever - HuggingFace transformers require mean pooling

Google-Gemini

Prajjwal1 - Converted Google BERT PyTorch model

Jonatasgrosman - Fine-tuned with OVHcloud GPU credits

Sentence-transformers - Maps text to 384-dimensional vectors

sentence Paraphrase - Text to 384D vectors, clustering, semantic search

Yiyanghkust - Pre-trained on financial text

WhiteRabbitNeo-33B

Ehsanaghaei - Specific RoBERTa for cybersecurity

Google-bert - English MLM pretrained model

Phind-CodeLlama-34B-v2

Tsmatz - Fine-tuned xlm-roberta-base

OpenVoice

Pyannote - Requires pyannote.audio v3.1+

Guillaumekln - OpenAI Whisper-large-v2 to CTranslate2 conversion

google/gemini-pro

Cardiffnlp - ~124M tweets, TweetEval sentiment finetuned.

sentence MiniLM - Sentence-embeddings: 384D, cluster, search

Facebook Encodec - EnCodec model card: Real-time audio codec

Microsoft - Pre-trained multimodal Transformer

Lxyuan - Distilled, zero-shot, Multilingual Sentiment

Openai Clip - Model card from CLIP repository

Cardiffnlp - XLM-roBERTa-base, ~198M tweets, sentiment finetuned

Microsoft - Improved BERT, RoBERTa with disentangled attention

Jean-Baptiste - Validated on emails/chat, outperformed others

Deepset - roberta-base, fine-tuned SQuAD2.0

Google Bert - Pretrained model Top 104 languages, MLM

Timm - MobileNet-v3 image classifier

tiiuae/falcon-7b

abacusai/Smaug-72B-v0.1

Facebook - Base model, 960h Librispeech, 16kHz fine-tuned

Cardiffnlp - roBERTa-base, 58M tweets, TweetEval irony finetuned

Genie by Lumalabs - text to 3d

Sentence Transformers - Maps text to 768D vectors

Cardiffnlp - roBERTa-base, ~58M tweets, TweetEval offensive language

Martin-ha - Toxic comments classification

Almanach - French RoBERTa-based model

Stabilityai - Weights for diffusers library

ai21labs/Jamba-v0.1

CAMeL Lab - Sentiment Analysis, fine-tuned

google - Vit - ImageNet-21k pre-trained, 224x224 resolution

Phi 2

Morpheus 1 - multi-modal generative ultrasonic transformer

Sentence Multi - Multi-QA MiniLM-L6, sentence-transformers

Google Electra - ELECTRA Self-supervised language learning.

Stabilityai diffusion - SDXL Ensemble, latent diffusion pipeline

google Fnetbase - English MLM, NSP pretrained model

Perplexity - PPLX Online

Gemma 7b by Google

Distilbert - Paper introduces distilled BERT

Runwayml - Stable Diffusion Text-to-image model

anthropic/claude-3-opus

Prajjwal1 - PyTorch model from converted Google BERT

Mixtral 8x7B

mixedbread-ai/mxbai-embed-large-v1

Dslim - Fine-tuned BERT, state-of-the-art NER

Ahmedrachid - Pre-trained on financial texts

Openai - AI explores vision robustness

BAAI - FlagEmbedding Retrieval-augmented LLMs

Timbrooks - Install diffusers in InstructPix2Pix using main for now

Supabase - GTE models by Alibaba DAMO Academy

LTP/small - Chinese NLP tools

Ashishkr - Validate content well-formedness

Stabilityai/stablelm-zephyr-3b

Sentence-transformers - 768-dimensional text vectors

Cmarkea - Fine-tuned for French NER.

Nlpconnect - Flax image captioning model, PyTorch version

Stabilityai - Fast text-to-image synthesis

Jonatasgrosman - OVHcloud GPU credits used

Deci/DeciLM-7B

Distilbert - Base multilingual model.

MythoMax-L2-13B-GPTQ

Bigscience/bloomz-560m

Sentence Qa - 768D dense vectors, semantic search

FMA-Net - video deblurring

SamLowe - roberta-base, go_emotions dataset, multi-label.

Facebook Opt-125m - Introduced in metaseq's repository

Facebook Sam - Object masks from input prompts

Jean Baptiste - Fine-tuned on wikinerfr

Microsoft Mdeberta - DeBERTa Improved BERT, RoBERTa

Papluca - XLM-RoBERTa with classification head

Distilbert Tokens - 768D vectors, clustering, semantic search

Facebook - BART-large checkpoint, trained on MNLI

Albert - English MLM pretrained model

Google-bert - English model, MLM objective

Cambridgeltl - SapBERT ACL 2021 cross-lingual extension

Google Patch - Pre-trained ImageNet-21k, fine-tuned ImageNet

Google Bert - Chinese model, input masking

deepset/gbert-large

Databricks - DBRX

Facebook Bart - Pre-trained English, fine-tuned CNN Daily Mail

Lmsys/vicuna-7b-v1.5

Jonatasgrosman - Fine-tuned wav2vec2-xl53: Russian, Common Voice, CSS10

Mistralai - Generative Sparse Mixture of Experts

CohereForAI/c4ai-command-r-v01-4bit

kk08 - Custom Crypto Market Sentiment

Microsoft Beit - Pre-trained ImageNet-22k, fine-tuned at 224x224.

Mistralai - Mistral-7B-Instruct-v0.1: Fine-tuned instruct LLM

Facebook xlm - CommonCrawl data, 100 languages

MAGNeT - Music generator by Meta AI

Open-Orca/Mistral-7B-OpenOrca

Tohoku - Japanese BERT pretrained model