Get a weekly rundown of the latest AI models and research... subscribe! https://aimodels.substack.com/

Declare-lab

Rank:

Average Model Cost: $0.0205

Number of Runs: 169,106

Models by this creator

šŸ’¬

flan-alpaca-gpt4-xl

The flan-alpaca-gpt4-xl model is a language model trained by OpenAI. It is part of the OpenAI GPT (Generative Pre-trained Transformer) series and represents the fourth iteration of the GPT model. GPT-4 XL is specifically designed to generate text based on the given prompts or instructions. It has been trained on a vast amount of text data and has achieved exceptional performance in various natural language processing tasks, such as text completion, summarization, translation, and question answering. With its advanced capabilities, this model can assist in automating content generation, conversational agents, and other applications that involve natural language understanding and generation.

Read more

$-/run

113.9K

Huggingface

šŸ¤”

tango

The Tango model uses instruction-guided diffusion to convert text into audio. It takes text input and generates coherent and natural-sounding audio output, using a combination of language and acoustic models. The instruction-guided diffusion technique allows the model to take into account additional guidance or instructions provided along with the text input, resulting in more accurate and customizable audio output. This model can be helpful in various applications such as text-to-speech systems, virtual assistants, and audio content generation.

Read more

$0.205/run

17.6K

Replicate

šŸ“Š

flan-alpaca-large

The Flan-Alpaca-Large model is a fine-tuned language model that specializes in problem-solving through instruction tuning. It is an extension of the Stanford Alpaca synthetic instruction tuning approach, specifically designed for existing instruction-tuned models like Flan-T5. The model has been evaluated on various open-source instruction-tuned language models and has shown improved performance. It can also be used for text-to-audio generation. The code, datasets, and pretrained models are available on the Hugging Face repository. The model aims to address the limitations and noise in the original Alpaca implementation by utilizing a fully accessible model trained on high-quality instructions.

Read more

$-/run

14.6K

Huggingface

šŸ“‰

flan-alpaca-base

The Flan-Alpaca model is a fine-tuned version of the Vicuna-13B model specifically designed for problem-solving tasks. It leverages the Alpaca approach to approximate the performance of large language models (LLMs) like ChatGPT by generating synthetic training data through instructions generated by an LLM. The Flan-Alpaca model is fully accessible and trained on high-quality instructions, making it a valuable resource for text generation and problem-solving tasks.

Read more

$-/run

12.8K

Huggingface

šŸ

flan-alpaca-xl

The Flan-Alpaca-XL model is an extension of the Stanford Alpaca synthetic instruction tuning approach to existing instruction-tuned models such as Flan-T5. It leverages large language models like GPT-3 to generate instructions as synthetic training data, which can then be used to fine-tune a smaller model. The Flan-Alpaca-XL model is fully accessible and trained on high-quality instructions, making it a useful tool for text-to-text generation tasks.

Read more

$-/run

5.4K

Huggingface

šŸŒ€

mustango

Controllable Text-to-Music Generation

Read more

$-/run

2.3K

Replicate

šŸ“¶

flan-alpaca-xxl

šŸ® šŸ¦™ Flan-Alpaca: Instruction Tuning from Humans and Machines šŸ“£ We developed Flacuna by fine-tuning Vicuna-13B on the Flan collection. Flacuna is better than Vicuna at problem-solving. Access the model here https://huggingface.co/declare-lab/flacuna-13b-v1.0. šŸ“£ Curious to know the performance of šŸ® šŸ¦™ Flan-Alpaca on large-scale LLM evaluation benchmark, InstructEval? Read our paper https://arxiv.org/pdf/2306.04757.pdf. We evaluated more than 10 open-source instruction-tuned LLMs belonging to various LLM families including Pythia, LLaMA, T5, UL2, OPT, and Mosaic. Codes and datasets: https://github.com/declare-lab/instruct-eval šŸ“£ FLAN-T5 is also useful in text-to-audio generation. Find our work at https://github.com/declare-lab/tango if you are interested. Our repository contains code for extending the Stanford Alpaca synthetic instruction tuning to existing instruction-tuned models such as Flan-T5. We have a live interactive demo thanks to Joao Gante! We are also benchmarking many instruction-tuned models at declare-lab/flan-eval. Our pretrained models are fully available on HuggingFace šŸ¤— : *recommended for better performance Why? Alpaca represents an exciting new direction to approximate the performance of large language models (LLMs) like ChatGPT cheaply and easily. Concretely, they leverage an LLM such as GPT-3 to generate instructions as synthetic training data. The synthetic data which covers more than 50k tasks can then be used to finetune a smaller model. However, the original implementation is less accessible due to licensing constraints of the underlying LLaMA model. Furthermore, users have noted potential noise in the synthetic dataset. Hence, it may be better to explore a fully accessible model that is already trained on high-quality (but less diverse) instructions such as Flan-T5. Usage

Read more

$-/run

1.7K

Huggingface

šŸ§ 

flan-gpt4all-xl

šŸ® šŸ¦™ Flan-Alpaca: Instruction Tuning from Humans and Machines šŸ“£ We developed Flacuna by fine-tuning Vicuna-13B on the Flan collection. Flacuna is better than Vicuna at problem-solving. Access the model here https://huggingface.co/declare-lab/flacuna-13b-v1.0. šŸ“£ Curious to know the performance of šŸ® šŸ¦™ Flan-Alpaca on large-scale LLM evaluation benchmark, InstructEval? Read our paper https://arxiv.org/pdf/2306.04757.pdf. We evaluated more than 10 open-source instruction-tuned LLMs belonging to various LLM families including Pythia, LLaMA, T5, UL2, OPT, and Mosaic. Codes and datasets: https://github.com/declare-lab/instruct-eval šŸ“£ FLAN-T5 is also useful in text-to-audio generation. Find our work at https://github.com/declare-lab/tango if you are interested. Our repository contains code for extending the Stanford Alpaca synthetic instruction tuning to existing instruction-tuned models such as Flan-T5. We have a live interactive demo thanks to Joao Gante! We are also benchmarking many instruction-tuned models at declare-lab/flan-eval. Our pretrained models are fully available on HuggingFace šŸ¤— : *recommended for better performance Why? Alpaca represents an exciting new direction to approximate the performance of large language models (LLMs) like ChatGPT cheaply and easily. Concretely, they leverage an LLM such as GPT-3 to generate instructions as synthetic training data. The synthetic data which covers more than 50k tasks can then be used to finetune a smaller model. However, the original implementation is less accessible due to licensing constraints of the underlying LLaMA model. Furthermore, users have noted potential noise in the synthetic dataset. Hence, it may be better to explore a fully accessible model that is already trained on high-quality (but less diverse) instructions such as Flan-T5. Usage

Read more

$-/run

428

Huggingface

šŸ›ø

segue-w2v2-base

Repository: https://github.com/declare-lab/segue Paper: https://arxiv.org/abs/2305.12301 SEGUE is a pre-training approach for sequence-level spoken language understanding (SLU) tasks. We use knowledge distillation on a parallel speech-text corpus (e.g. an ASR corpus) to distil language understanding knowledge from a textual sentence embedder to a pre-trained speech encoder. SEGUE applied to Wav2Vec 2.0 improves performance for many SLU tasks, including intent classification / slot-filling, spoken sentiment analysis, and spoken emotion classification. These improvements were observed in both fine-tuned and non-fine-tuned settings, as well as few-shot settings. How to Get Started with the Model To use this model checkpoint, you need to use the model classes on our GitHub repository. You do not need to create the Processor yourself, it is already available as model.processor. SegueForRegression and SegueForClassification are also available. For classification, the number of classes can be specified through the n_classes field in model config, e.g. SegueForClassification.from_pretrained('declare-lab/segue-w2v2-base', n_classes=7). Multi-label classification is also supported, e.g. n_classes=[3, 7] for two labels with 3 and 7 classes respectively. Pre-training and downstream task training scripts are available on our GitHub repository. Results We show only simplified MInDS-14 and MELD results for brevity. Please refer to the paper for full results. MInDS-14 (intent classification) Note: we used only the en-US subset of MInDS-14. Note: Wav2Vec 2.0 fine-tuning was unstable. Only 3 out of 6 runs converged, the result shown were taken from converged runs only. MELD (sentiment and emotion classification) Note: Wav2Vec 2.0 fine-tuning was unstable at the higher LR. Limitations In the paper, we hypothesized that SEGUE may perform worse on tasks that rely less on understanding and more on word detection. This may explain why SEGUE did not manage to improve upon Wav2Vec 2.0 on the Fluent Speech Commands (FSC) task. We also experimented with an ASR task (FLEURS), which heavily relies on word detection, to further demonstrate this. However, this is does not mean that SEGUE performs worse on intent classification tasks in general. MInDS-14, was able to benifit greatly from SEGUE despite also being an intent classification task, as it has more free-form utterances that may benefit more from understanding. Citation

Read more

$-/run

254

Huggingface

šŸ—£ļø

tango-full-ft-audiocaps

Platform did not provide a description for this model.

Read more

$-/run

178

Huggingface

Similar creators