Guides
- Quickstart
- Quickstarts by language
- Migration guides
- Authentication and keys
- Key protection
- Models, aliases and fallbacks
- Automatic model choice (silicon/auto)
- Chat completions
- Streaming
- Structured outputs and tools
- Logprobs
- Vision
- Responses, Claude and Ollama
- Embeddings, rerank, audio, images
- Files and batches
- Documents and search
- LoRA adapters
- Project tags and costs
- Prefix caching
- Headers, rate limits and spend caps
- Errors
- Versioning and the sandbox
- Privacy and retention
- Verifiable receipts
- Data residency
- Pricing
- Billing
- Pay per request (x402)
- On-chain oracle
- Agent credit (backed compute lines)
- Status
- SDKs
- Swift SDK
- Integrations
- Run a node
- Linux agent
- Mac mini fleets
- Compute API
- Architecture
- Changelog
- Glossary
API reference
Guide
Quickstarts by language
Pick your language. Each page takes about five minutes: install, set the key, one chat completion, a streamed answer, the model list and an error handled.
Before you start
Create an API key on the API Keys page and add balance on the Billing page. Every client talks to https://api.siliconnetwork.io/v1 with the key as a bearer token. The Python, TypeScript, Go and Rust clients read the key from SILICON_API_KEY; the Swift client takes it as an argument.
The examples use llama-3.2-3b. GET /v1/models lists what is loaded right now; the routing alias fast picks whichever small model is serving.
Languages
| Language | Package | Page |
|---|---|---|
| Python | siliconnetwork | Python quickstart |
| TypeScript | @siliconnetwork/sdk | TypeScript quickstart |
| Go | sdks/go (from the repository) | Go quickstart |
| Rust | siliconnetwork (from the repository) | Rust quickstart |
| Swift | SiliconNetwork (local package) | Swift quickstart |
| Any (HTTP) | curl | curl quickstart |
The Go, Rust and Swift clients are not on a registry yet, so their pages say how to use them from a checkout of the repository; the Python and TypeScript pages say the same for when a package is not on PyPI or npm yet. Already on the OpenAI, Anthropic or Ollama clients? Keep them and change the base URL: see the migration guides.