Compare Models
-
Microsoft
Azure OpenAI Service
OTHERMicrosoft’s Azure OpenAI Service allows you to take advantage of large-scale, generative AI models with deep understandings of language and code to enable new reasoning and comprehension capabilities for building cutting-edge applications. Apply these coding and language models to a variety of use cases, such as writing assistance, code generation, and reasoning over data. Detect and mitigate harmful use with built-in responsible AI and access enterprise-grade Azure security. GPT-4 is available in preview in the Azure OpenAI Service and the billing for GPT-4 8K and 32K instances per 1/K tokens and can be found under those models on the tokes compare site. To note, Microsoft’s Azure OpenAI Service customers can access GPT-3.5, ChatGPT, and DALL·E too. -
Google
BARD
FREEGoogle’s Bard is now powered by PaLM 2, the new powerful LLM launched in May 2023. PaLM 2 is trained on a massive dataset of text and code. Bard can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way. Bard is programmed to use the web to find the most recent answers to questions. This means that when you ask Bard a question, it will not only use its knowledge of the world to answer your question, but it will also use the internet to find the most recent information on the topic. This allows Bard to provide you with the most accurate and up-to-date information possible (very cool).The exact billing structure for Bard is still under development (it is free to try at the moment) but you will likely be able to purchase tokens in bulk at a discounted price. According to Google, you may also be able to use tokens you have earned through other means, such as completing surveys or participating in beta testing programs. -
Google
BERT
FREEBERT (Bidirectional Encoder Representations from Transformers) was introduced in 2018 by researchers at Google AI. BERT uses AI in the form of natural language processing (NLP), natural language understanding (NLU), and sentiment analysis to process every word in a search query in relation to all the other words in a sentence, giving it a robust understanding of context and semantics. This pre-training process is incredibly powerful and the learned weights can be fine-tuned with just one additional output layer to create models for a variety of NLP tasks such as question answering and sentiment analysis. You can download the smaller BERT models for FREE from the official BERT GitHub page. -
Microsoft
Bing Search APIs
OTHERMicrosoft’s Bing AI search engine is powered by GPT-4. Microsoft claims the new model is faster and more accurate than ever. Bing Search APIs provide a variety of APIs with trained models for your use. The Bing Search APIs add intelligent search to your app, combining hundreds of billions of webpages, images, videos, and news to provide relevant results without ads. The results can be automatically customized to your user’s locations or markets, increasing relevancy by staying local. There are various prices for Bing Search APIs which are dependent on the feature. For customers who are interested in more flexible terms related to presenting Bing API results with their models check out the website for prices per 1,000 transactions. -
BigScience
BLOOM
FREEBigScience Large Open-science Open-access Multilingual Language Model (BLOOM) is a transformer-based LLM. Over 1,000 AI researchers created it to provide a free large language model for everyone who wants to try and it is a multilingual LLM. BLOOM is an autoregressive Large Language Model (LLM), trained to continue text from a prompt on vast amounts of text data using industrial-scale computational resources. It can output coherent text in 46 languages and 13 programming languages. It is free, and everybody who wants to can try it out. To interact with the API, you’ll need to request a token. This is done with a post request to the server. Tokens are only valid for two weeks. After which, a new one must be generated. Trained on around 176B parameters, it is considered an alternative to OpenAI models. There is a downloadable model, and a hosted API is available. -
ChatGLM
ChatGLM-6B
FREEResearchers at the Tsinghua University in China have worked on developing the ChatGLM series of models that have comparable performance to other models such as GPT-3 and BLOOM. ChatGLM-6B is an open bilingual language model (trained on Chinese and English). It is based on General Language Model (GLM) framework, with 6.2B parameters. With the quantization technique, users can deploy locally on consumer-grade graphics cards (only 6GB of GPU memory is required at the INT4 quantization level). The following models are available: ChatGLM-130B (an open source LLM), ChatGLM-100B (not open source but available through invite-only access), and ChatGLM-6 (a lightweight open source alternative). ChatGLM LLMs are available with a Apache-2.0 license that allows commercial use. We have included the link to the Hugging Face page where you can try the ChatGLM-6B Chatbot for free. -
Deepmind
Chinchilla AI
OTHERGoogle’s DeepMind Chinchilla AI is still in the testing phase. Once released, Chinchilla AI will be useful for developing various artificial intelligence tools, such as chatbots, virtual assistants, and predictive models. It functions in a manner analogous to that of other large language models such as GPT-3 (175B parameters), Jurassic-1 (178B parameters), Gopher (280B parameters), and Megatron-Turing NLG (300B parameters) but because Chinchilla is smaller (70B parameters), inference and fine-tuning costs less, easing the use of these models for smaller companies or universities that may not have the budget or hardware to run larger models.
-
Google
Cloud Platform
OTHERGoogle Cloud Platform (GCP) is a cloud computing service that includes innovative AI and machine learning products, solutions, and services. Google AI Studio is a low-code development environment that makes it easy to build and deploy applications and has a variety of features, such as pre-trained models that can be used to get started quickly, a unified experience for managing the entire ML lifecycle, from data preparation to model deployment, and a variety of tools for monitoring the performance of ML models in production. Vertex AI can be used to train and deploy models, and GCP also offers a variety of data storage services, including Cloud Storage, which can be used to store large datasets. -
Google
code chat (codechat-bison)
$0.002Based on Google’s PaLM 2 large language model, the company specifically trained Codey APIs to handle coding-related prompts, but it also trained the model to handle queries related to Google Cloud.
The code chat API can power a chatbot that assists with code-related questions. For example, you can use it for help debugging code. The code chat API supports the code-chat-bison model.
The Codey APIs support a wide range of programming languages, including C++, C#, Go, GoogleSQL, Java, JavaScript, Kotlin, PHP, Python, Ruby, Rust, Scala, Swift, and TypeScript. You can run with the API and in Generative AI Studio.
Some common use cases for code chat include debugging, where it assists with issues related to code that doesn’t compile or contains a bug; documentation, where it aids in understanding unfamiliar code to ensure accurate representation; and learning, as it provides help in comprehending code that you might not be very familiar with.
Note: We have converted characters to tokens for the prices (based on the approximation of 4 characters per 1 token).
-
Google
code completion (code-gecko)
$0.002Based on Google’s PaLM 2 large language model, the company specifically trained Codey APIs to handle coding-related prompts, but it also trained the model to handle queries related to Google Cloud. The code completion API provides code autocompletion suggestions as you write code. The API uses the context of the code you’re writing to make its suggestions.
The code completion API supports the code-gecko model. Use the code-gecko model to help improve the speed and accuracy of writing code. The Codey APIs support a wide range of programming languages including C++, C#, Go, GoogleSQL, Java, JavaScript, Kotlin, PHP, Python, Ruby, Rust, Scala, Swift, and TypeScript. You can run with the API and in Generative AI Studio. Some common use cases for code completion include writing code faster, where the code-gecko model is employed to expedite the coding process by leveraging suggested code; and minimizing bugs in code, by utilizing code suggestions that are known to be syntactically correct to circumvent errors, thus reducing the risk of inadvertently introducing bugs that can arise during code creation.
Note: We have converted characters to tokens for the prices (based on the approximation of 4 characters per 1 token).
-
Google
code generation (code-bison)
$0.002Based on Google’s PaLM 2 large language model, the company specifically trained Codey APIs to handle coding-related prompts, but it also trained the model to handle queries related to Google Cloud.
code generation (code-bison) generates code based on a natural language description of the desired code. For example, it can generate a unit test for a function. The code generation API supports the code-bison model. The Codey APIs support a wide range of programming languages, including C++, C#, Go, GoogleSQL, Java, JavaScript, Kotlin, PHP, Python, Ruby, Rust, Scala, Swift, and TypeScript. You can run with the API and in Generative AI Studio.
Some common use cases for code generation include creating unit tests, where you can design a prompt to request a unit test for a specific function; writing a function, which involves passing a problem to the model and receiving a function that solves the problem; and creating a class, where you can use a prompt to describe the purpose of a class and have the code defining that class returned to you.
Note: We have converted characters to tokens for the prices (based on the approximation of 4 characters per 1 token).
-
Google, Stanford University
Electra
FREEELECTRA (Efficiently Learning an Encoder that Classifies Token Replacements Accurately) is a transformer-based model like BERT, but it uses a different pre-training approach, which is more efficient and requires less computational resources. It was created by a team of researchers from Google Research, Brain Team, and Stanford University. ELECTRA models are trained to distinguish “real” input tokens vs “fake” input tokens generated by another neural network (for the more technical audience, ELECTRA uses a new pre-training task, called replaced token detection (RTD), that trains a bidirectional model while learning from all input positions). Inspired by generative adversarial networks (GANs), ELECTRA trains the model to distinguish between “real” and “fake” input data. At small scale, ELECTRA achieves strong results even when trained on a single GPU. At large scale, ELECTRA achieves state-of-the-art results on the SQuAD 2.0 dataset. Go to GitHub where you can access the three models (ELECTRA-Small, ELECTRA-Base and ELECTRA-Large).
