What is an alternative to Openai api?
Anil Matcha
7 replies
Interested to know the possibilities
Replies
Fabien Snauwaert@fabiensnauwaert
Gliglish
If you mean OpenAI's GPT models and large language models (LLMs), then the big alternatives are:
GPT-Neo, GPT-J, GPT-NeoX which are all open source.
There are also projects from big companies, these come to mind:
- NeMo from Nvidia
- YaLM from Yandex
Then you need to find either a cloud service that offers them or have access to a GPU to run them.
Share
@fabiensnauwaert Yep, you could host the open source GPTs on AWS yourself, but that's a massive pain lol.
We do plan to have them on evoke-app.com as an API eventually though, so people would skip that process.
Unbody
Depends on what functionality you are looking for, if it is something you can solve using embeddings and vector search, for example, there is weaviate.io. Recently they added also generative models to their engine.
What are you looking for?
For image generation, I'd recommend a stable diffusion API like evoke-app.com
For LLMs, you don't really have many choices. There is GPT-NEOX, which is the closest to GPT-3 you can run.
We hope to get GPT-NEOX on Evoke in the near future as well as an API too
Microlaunch
OpenAI API mainly helps with:
- generating text from prompts
- generating code snippets
- generating images using Dall-e model
Some interesting alternatives:
- Midjourney: for state of the art image generation
- All APIs providing LLM training/prediction: Chatsonic, Jasper, Rytr
- Hugginface exposes a lot of LLMs
- Github copilot for code snippets
Interested in a full list, curated my own one on gumroad, 100% open
https://3912310453273.gumroad.co...
AskMiku
There are several alternatives to the OpenAI API, including:
Google Cloud AI Platform: Google Cloud AI Platform provides a suite of tools for developing and deploying machine learning models, including pre-built models, data labeling, and custom model training.
Amazon SageMaker: Amazon SageMaker is a fully-managed machine learning service that provides tools for building, training, and deploying machine learning models at scale.
Microsoft Azure Machine Learning: Microsoft Azure Machine Learning is a cloud-based machine learning service that provides tools for building, training, and deploying machine learning models. It also includes pre-built models and data labeling tools.
IBM Watson Studio: IBM Watson Studio is a cloud-based platform for building, training, and deploying machine learning models. It includes tools for data preparation, model development, and deployment.
Hugging Face: Hugging Face is an open-source library of pre-trained models for natural language processing (NLP) and other machine learning tasks. It also provides a platform for training and deploying custom models.
These alternatives provide similar functionality to the OpenAI API and can be used for a wide range of machine learning tasks. The choice of which platform to use will depend on your specific needs and requirements.
Anthropic AI and Character AI