If you're looking for a platform that provides low-cost and high-performance NLP models for text classification, sentiment analysis, and entity recognition, NuMind could be a good option for you. NuMind is a machine learning platform that lets you create text processing models without writing any code. It supports multilingual text classification, sentiment analysis, and entity recognition, as well as other tasks. The platform is designed to be data efficient and cost effective, with a Hobbyist plan that's free and an Enterprise plan for custom needs.
Another option is Predibase, a developer-focused service for fine-tuning and serving large language models (LLMs) at a low cost. It's good for classification, information extraction and code generation. Predibase offers free serverless inference for up to 1 million tokens per day and a pay-as-you-go pricing model that charges based on model size and the size of the data you're using, so it's good for small and large scale NLP work.
If you want to tap into a broad library of AI models through a single API, AIML API is an option. The service has more than 100 AI models and includes serverless inference so you can easily add advanced machine learning abilities to your projects. Pricing starts at $0.45 per million tokens, so it's relatively cheap and scalable.
Last, MonkeyLearn is a no-code text analytics tool that gives you immediate access to customer feedback insights with customizable charts and graphs. It offers pre-trained and custom machine learning models for text classification and data extraction, and integrates with many apps and business intelligence tools. It's good for automating text analysis and getting insights without having to write a lot of code.