Text inference
Get responses from text models through an API: write, explain, translate and process text.
Network achievements: what is unlocked, what is worth trying and what is still ahead.
Two ways to participate: use the models or contribute compute to the network.
Get responses from text models through an API: write, explain, translate and process text.
Connect supported hardware, become a host and participate in the network and its reward system.
Practical use cases for text models. Each status helps you choose where to start.
Process text through the API: classify, extract data and create short summaries.
Use a model in your research workflow. Check facts and conclusions against original sources.
Discuss ideas, ask questions and work with text in a familiar chat interface.
Embed text models in an application or bot. Evaluate the fit using your own workflows.
Code assistance in your editor: explanations, edits and working with a project.
Multi-step tasks, tool use and independently carrying out actions.
The next achievements: new capabilities and more predictable network performance.
Running your own training or fine-tuning is not yet a ready-to-use public network service.
A ready-to-use service that hides request content from compute hosts is not yet available.
Generating images through network models is not yet available.
Generating video, speech, music and other audio through network models is not yet available.
For every task: fewer failures, shorter delays and fewer problems with model responses.