Models

OneContext currently supports all sentence-transformer models. That's (at time of writing) 4,402 models. You can browse the model-cards and compare performance on benchmarks here.

If you are on the enterprise plan, your selected models will be attached to a private instance of OneContext, and will always be attached to dedicated GPUs. If you are on the starter plan, your selected models will be attached to a shared instance of OneContext, and will be attached to a shared pool of GPUs.

For more on sentence-transformers, check out the documentation here, and here.

If you'd like to use a model not supported by sentence-transformers, or, would like us to host and deploy a model of your own (for example, a model you have fine-tuned), please reach out, and we'll try to help.

Alternatively, if you'd like us to fine-tune your model for you, please reach out. This is not something we currently support, but we are open to thinking about doing this in the future if there is enough demand!