Hosted open-weight models via REST and official Python / TypeScript SDKs
Together AI operates a developer platform for running prominent open-source and vendor-weight models from Together-hosted GPUs. Documentation centers on issuing API keys, installing the Together Python (`together`) or npm (`together-ai`) SDKs, or calling HTTPS endpoints such as `https://api.together.ai/v1/chat/completions` with Bearer authentication. Guides cover streaming chat completions, function calling, structured outputs, model catalog browsing, GPU reservations for steady traffic, and fine-tuning or dedicated cluster offerings published in the broader docs hierarchy.
Use cases
- Shipping chat or agent backends on hosted open-weight checkpoints without owning GPU fleets
- Prototyping tool-calling pipelines using Together-supported models listed in docs
- Moving from curl-only proofs to typed SDK integrations for retries and telemetry
- Exploring Together fine-tuning or dedicated endpoint SKUs once baseline usage patterns are validated
- Teaching teams an OpenAI-shaped HTTP interface while swapping model IDs to Together catalog entries
Key features
- Official SDK quickstart flows for Python and TypeScript with environment-based API keys (`TOGETHER_API_KEY`)
- REST chat-completions endpoints compatible with common OpenAI-style JSON payloads illustrated in Together quickstart docs
- Streaming completions demonstrated with `stream=True` in Python and Async iterators in TypeScript samples
- Product surface area spanning model catalogs, GPU clusters, LoRA/full fine-tuning, and reservations per Together documentation index
- Documented linkage between keys, billing projects (`api.together.ai` console), and per-model selection
Who Is It For?
- Backend engineers prototyping LLM-heavy services
- ML engineers evaluating hosted inference for open-weight checkpoints
- Developer advocates standardizing onboarding material with official SDK snippets
Frequently Asked Questions
- Do I need the SDK or can I use curl?
- Together documents both: quickstart installs `together` or `together-ai`, but curl examples POST directly to `/v1/chat/completions` with your API key header.
- Where are API keys created?
- Docs direct users to the Together console API keys workflow under their active project (`api.together.ai/settings/projects/~current/api-keys` per quickstart).
- Which model does the introductory sample call?
- The quickstart streams `openai/gpt-oss-20b` as of the Together quickstart reproduction in this corpus.
Related
Related
3 Indexed items
Replicate
Replicate is a hosted platform for executing third-party and custom machine-learning models over HTTP without provisioning GPUs yourself. Official documentation explains how to authenticate with API tokens, create asynchronous predictions, stream outputs, retrieve model metadata, wire webhooks for completion events, and optionally deploy or fine-tune checkpoints (for example FLUX image workflows) published to the Replicate catalog.
Groq Cloud API
GroqCloud exposes hosted language, speech, and compound workloads through Groq’s HTTP APIs. Documentation highlights compatibility with OpenAI client libraries when you point `base_url` at Groq’s OpenAI-compatible endpoint and supply a Groq API key, alongside first-party Groq SDKs for Python and JavaScript. Pricing pages publish per-model token rates (USD) for on-demand inference.
Hugging Face Hub
Hugging Face operates the Hugging Face Hub—a central place to browse and host machine-learning artifacts—alongside Spaces for demo apps and documentation for calling models through HTTP APIs using Hugging Face access tokens. Official docs outline creating accounts and tokens (`Settings → Access Tokens`), downloading files with Git LFS-compatible clients, versioning repositories, and invoking models through Inference Providers / serverless patterns published in huggingface.co documentation rather than stitching together bespoke hosting.