Implementation:Togethercomputer Together python Models Resource

Knowledge Sources	Together Python
Domains	Model_Management, API
Last Updated	2026-02-15 16:00 GMT

Overview

Concrete tool for listing available models and uploading custom models to the Together AI platform provided by the Together Python SDK.

Description

The Models class provides list() (with optional dedicated-only filter) and upload() methods. Listing returns all available models sorted alphabetically with metadata including type, pricing, context length, and organization. Upload supports custom models and adapters from Hugging Face repos or S3 paths.

Usage

Import this class when you need to discover available models or upload custom models/adapters. Access via client.models.

Code Reference

Source Location

Repository: Together Python
File: src/together/resources/models.py
Lines: 1-252

Signature

class Models(ModelsBase):
    def list(self, dedicated: bool = False) -> List[ModelObject]: ...

    def upload(
        self,
        *,
        model_name: str,
        model_source: str,
        model_type: str = "model",
        hf_token: str | None = None,
        description: str | None = None,
        base_model: str | None = None,
        lora_model: str | None = None,
    ) -> ModelUploadResponse: ...

Import

from together import Together

client = Together()
# Access via client.models

I/O Contract

Inputs

Name	Type	Required	Description
dedicated	bool	No	If True, return only dedicated-deployable models (default: False)
model_name	str	Yes (upload)	Name for the uploaded model
model_source	str	Yes (upload)	HuggingFace repo or S3 path
model_type	str	No	"model" or "adapter" (default: "model")
hf_token	str	No	HuggingFace access token
base_model	str	No	Base model for adapter type

Outputs

Name	Type	Description
list() returns	List[ModelObject]	Sorted list of models with id, type, pricing, context_length
upload() returns	ModelUploadResponse	Upload job info: job_id, model_name, model_id, message

Usage Examples

from together import Together

client = Together()

# List all models
models = client.models.list()
for model in models:
    print(f"{model.id}: {model.type} (ctx: {model.context_length})")

# List only dedicated-deployable models
dedicated = client.models.list(dedicated=True)

# Upload a custom model from HuggingFace
response = client.models.upload(
    model_name="my-org/my-custom-model",
    model_source="huggingface/my-model-repo",
    model_type="model",
    hf_token="hf_...",
    description="My fine-tuned model",
)
print(f"Upload job: {response.job_id}")

# Upload a LoRA adapter
response = client.models.upload(
    model_name="my-org/my-adapter",
    model_source="huggingface/my-adapter-repo",
    model_type="adapter",
    base_model="meta-llama/Llama-4-Scout-17B-16E-Instruct",
)

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment