# Webway

> Placeholder description for Webway. Replace this with one sentence describing what the gateway does — it becomes the meta description sitewide.

Webway is an OpenAI-compatible gateway: one endpoint routes completions across providers, meters tokens, and bills against prepaid credit. The catalogue at https://alpha.webway.sh is public; each model has its own page with pricing and context window.

## Models

- [Claude Opus 5 (Anthropic)](https://alpha.webway.sh/models/claude-opus-5): Frontier Claude model for hard reasoning, large codebases, and agentic work. Context 200K; $1.10 per 1M input tokens, $5.50 per 1M output.
- [Claude Sonnet 5 (Anthropic)](https://alpha.webway.sh/models/claude-sonnet-5): Cost-efficient Claude tier for high-volume coding, chat, and tool use. Context 200K; $3.00 per 1M input tokens, $15.00 per 1M output.
- [GPT-5.6 Luna (OpenAI)](https://alpha.webway.sh/models/gpt-5.6-luna): Cheapest GPT-5.6 tier, for high-volume chat, classification and light agentic work. Context 1.1M; $0.20 per 1M input tokens, $1.20 per 1M output.
- [GPT-5.6 Sol (OpenAI)](https://alpha.webway.sh/models/gpt-5.6-sol): Flagship GPT-5.6 tier, for long-horizon agentic work, hard reasoning and command-line coding. Context 1.1M; $4.00 per 1M input tokens, $20.00 per 1M output.
- [GPT-5.6 Terra (OpenAI)](https://alpha.webway.sh/models/gpt-5.6-terra): Mid GPT-5.6 tier, balancing capability against cost for everyday coding, reasoning and tool use. Context 1.1M; $2.00 per 1M input tokens, $12.00 per 1M output.
- [DeepSeek V4 Flash (DeepSeek)](https://alpha.webway.sh/models/deepseek-v4-flash): DeepSeek's fast tier, for high-volume work that still wants the million-token window. Context 1M; $0.44 per 1M input tokens, $1.32 per 1M output.
- [DeepSeek V4 Pro (DeepSeek)](https://alpha.webway.sh/models/deepseek-v4-pro): DeepSeek's frontier tier, a million-token window at a fraction of frontier prices. Context 1M; $1.32 per 1M input tokens, $3.96 per 1M output.
- [Kimi K2.7 Code (Moonshot AI)](https://alpha.webway.sh/models/kimi-k2.7-code): Kimi's dedicated coding model, for instruction-following across a whole repository. Context 262K; $0.95 per 1M input tokens, $4.00 per 1M output.
- [Kimi K3 (Moonshot AI)](https://alpha.webway.sh/models/kimi-k3): Moonshot's frontier model, a million-token window for software engineering and deep reasoning. Context 1.0M; $3.00 per 1M input tokens, $15.00 per 1M output.

## Documentation

- [Introduction](https://alpha.webway.sh/docs): Webway is one OpenAI-compatible endpoint in front of every provider, billed per token from prepaid credit.
- [Quickstart](https://alpha.webway.sh/docs/quickstart): Mint a key, send a request, and see it land in usage — about two minutes end to end.
- [Your first request](https://alpha.webway.sh/docs/first-request): The chat completions body field by field, what comes back, and the two headers worth knowing about.
- [Authentication](https://alpha.webway.sh/docs/authentication): How keys are minted, stored and revoked, and why the raw value is shown exactly once.
- [Models](https://alpha.webway.sh/docs/models): What the catalogue holds, how a model is priced, and what happens when one is withdrawn.
- [Credits and billing](https://alpha.webway.sh/docs/concepts/credits): Prepaid credit, drawn down per request, exact to the micro-dollar — and the ledger that records every movement.
- [Usage and metering](https://alpha.webway.sh/docs/concepts/usage): One row per request — tokens, cost, latency and status — and what the dashboard builds out of it.
- [Zero data retention](https://alpha.webway.sh/docs/concepts/zero-data-retention): What the gateway keeps, what it never keeps, and the header that refuses a provider which would.
- [Streaming](https://alpha.webway.sh/docs/guides/streaming): Server-sent events, why time to first token is the number that matters, and how streamed calls are metered.
- [Auto-reload](https://alpha.webway.sh/docs/guides/auto-reload): A standing instruction to buy credit when the balance runs low, so requests do not start failing at 3am.
- [Bring your own key](https://alpha.webway.sh/docs/guides/bring-your-own-key): Route through your own provider account, keep your negotiated rates, and still get one endpoint and one usage view.
- [Teams](https://alpha.webway.sh/docs/guides/teams): Invite people into the organization that holds the credit, and give them the access their job needs.
- [Chat completions](https://alpha.webway.sh/docs/reference/chat-completions): POST /v1/chat/completions — parameters, headers, and the response object.
- [Errors](https://alpha.webway.sh/docs/reference/errors): Status codes, the error body, and which failures are worth retrying.
- [Rate limits](https://alpha.webway.sh/docs/reference/rate-limits): What is limited, what happens when you cross it, and how to find out before your users do.

## Blog

- [Hello world](https://alpha.webway.sh/blog/hello-world): A second placeholder post, here so the blog index and RSS feed have more than one entry.
- [Writing a post](https://alpha.webway.sh/blog/writing-a-post): Placeholder post explaining the frontmatter a blog entry needs. Replace before launch.

## Optional

- [Full text of every page](https://alpha.webway.sh/llms-full.txt): All docs and posts as plain markdown
- [RSS feed](https://alpha.webway.sh/blog/rss.xml): New blog posts