Multi-Modal Foundation · biohub
Experimental ESMFold2 variant exploring new training recipes and architectural changes for protein structure prediction, design, and confidence estimation.
< 1s
Runs on On-demand GPU
Runs on our servers with on-demand compute. A first run needs time to load the model; active capacity can be reused and scales down when idle.Run this model
About this model
Experimental ESMFold2 variant exploring new training recipes and architectural changes for protein structure prediction, design, and confidence estimation. Served through the generic Modal runner, which pulls biohub/ESMFold2-Experimental from the Hugging Face Hub and runs it as a feature-extraction pipeline. Inputs and outputs were derived from the repository's declared pipeline tag rather than written by hand — check the model card before relying on a result.
Standardized I/O contract
Every model in the Hub speaks the same contract, which is what lets the Router and the agent call any of them without special-casing.
Inputs
Sequence or text
Outputs
Mean-pooled representation, one vector per input.
