Models
Cutting-edge large language models are being unveiled at an unprecedented pace, leading to dynamic shifts on the various performance leaderboards. One of the core benefits of Continual is the ability to seamlessly switch be between different LLMs whether it's to optimize cost, enhance performance, or minimize downtime.
Continual supports many models out-of-the-box and Continual admins can add new models to their workspace by creating a new Model Provider in the Continual console. The new Model Provider must be compatible with one of the supported model provider types. For example, if the new model provider offers an OpenAI-compatible API then they can be added to Continual along with their models.
Default Large Language Models
NAME | PROVIDER | MODELS |
---|---|---|
Default Azure Open AI | azure | azure-gpt-35-turbo, azure-gpt4o |
Default Open AI | openai | gpt-4, gpt-3.5-turbo-0125, gpt-4-1106-preview, gpt-4-turbo-2024-04-09, gpt-4o |
Default Anthropic | anthropic | claude-3-opus-20240229, claude-3-sonnet-20240229, claude-3-haiku-20240307 |
Default Anyscale | anyscale | mistralai/Mixtral-8x7B-Instruct-v0.1, mistralai/Mixtral-8x22B-Instruct-v0.1 |
Default Databricks | databricks | databricks-dbrx-instruct, databricks-meta-llama-3-70b-instruct, databricks-llama-2-70b-chat, databricks-mixtral-8x7b-instruct |
Please note that some models do not support function calling which is necessary for Tool use.