Get a supported model
Retrieves a Together-hosted base model and the certified model, configuration, hardware, and performance profiles available for deployment.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Path Parameters
Supported model identifier.
Response
OK
Curated catalog entry for a platform-supported model.
Unique ID of the deployable Together-hosted base model.
Catalog-controlled HF model ID used for inference.
Catalog-controlled human-readable model name.
Input modalities supported by the model.
MODALITY_TEXT, MODALITY_IMAGE, MODALITY_AUDIO, MODALITY_VIDEO Output modalities produced by the model.
MODALITY_TEXT, MODALITY_IMAGE, MODALITY_AUDIO, MODALITY_VIDEO Product surfaces where the model is offered.
PRODUCT_SERVERLESS, PRODUCT_DEDICATED, PRODUCT_FINE_TUNING High-level tasks the model supports.
CAPABILITY_CHAT, CAPABILITY_EMBEDDING, CAPABILITY_RERANKING, CAPABILITY_IMAGE_GENERATION, CAPABILITY_VIDEO_GENERATION Organization or publisher associated with the model.
Catalog recommendation status for the model.
SUPPORTED_MODEL_STATUS_RECOMMENDED, SUPPORTED_MODEL_STATUS_SUPPORTED, SUPPORTED_MODEL_STATUS_DEPRECATED, SUPPORTED_MODEL_STATUS_HIDDEN UI-facing model type badge, such as chat, language, code, image, embedding, rerank, moderation, audio, video, or transcribe.
Bare model ID for the architecture's base model; empty when no base model is linked.
Resource name for the base model as projects/{projectId}/models/{modelId}; empty when unresolved.
Certified deployment profiles available for the model.
Timestamp when the catalog entry was created.
Timestamp when the catalog entry was last updated.
Human-readable model description.
Advanced features exposed by the model.
FEATURE_TOOL_CALLING, FEATURE_STRUCTURED_OUTPUT, FEATURE_REASONING Model architecture from the underlying weights metadata.
Maximum context length from the underlying weights metadata.
Searchable catalog tags for the model.
Preferred input format for the model.
Preferred output format for the model.
Serverless endpoint name for inference, if available.
Model family identifier for related catalog entries.