`together://`¶

Together AI — hosted inference for a wide catalogue of open-source models (Llama, Mixtral, Qwen, FLUX, Stable Diffusion, etc.) via an OpenAI-compatible API.

Built on pkg/component/llmutil/ — the same chat/embeddings shape as openai://. This page covers only the Together-specific defaults.

URI shape¶

together://chat?model=<id>&apiKey=<key>&temperature=<float>&maxTokens=<int>&baseURL=<url>
together://embeddings?model=<id>&apiKey=<key>

Parameters¶

Param	Type	Default	Producer/Consumer/Both	Description
path	enum	required	Producer	`chat` or `embeddings`.
`model`	string	`meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo`	Producer	Together model id — note Together's ids are typically `<org>/<model>` slashed. See https://docs.together.ai/docs/inference-models.
`apiKey`	string	(from constructor)	Producer	Together API key. Sent as `Authorization: Bearer <key>`.
`baseURL`	string	`https://api.together.xyz/v1`	Producer	Override for proxies.
`temperature`	float	`0.7`	Producer	Sampling temperature.
`maxTokens`	int	`4096`	Producer	Output cap.

Producer-only. Default timeout 120 s.

Headers¶

Namespaced under together.* (together.model, together.id, together.finishReason, together.usage.*). Input header systemPrompt works the same way.

Examples¶

- to: together://chat?model=meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo&apiKey=${TOGETHER_API_KEY}

Qwen 2.5 72B for non-English work:

- to: together://chat?model=Qwen/Qwen2.5-72B-Instruct-Turbo&apiKey=${TOGETHER_API_KEY}

Embeddings:

- to: together://embeddings?model=BAAI/bge-large-en-v1.5&apiKey=${TOGETHER_API_KEY}

Behavior notes¶

See openai://. Together's -Turbo variants are quantised — pick the non-turbo id when you need full precision.

Source¶

pkg/component/together/together.go.
pkg/component/llmutil/llmutil.go.

together://¶