diff --git a/docs/customize/model-providers/top-level/openai.mdx b/docs/customize/model-providers/top-level/openai.mdx index 4d7326b479e..706b6ef1661 100644 --- a/docs/customize/model-providers/top-level/openai.mdx +++ b/docs/customize/model-providers/top-level/openai.mdx @@ -165,3 +165,30 @@ By default, Continue uses OpenAI's `/responses` endpoint for o-series and gpt-5 ``` + +### Example: Kimi K3.1 Pro via Wallaby + +A tested configuration for running Kimi K3.1 Pro (Moonshot AI) through Wallaby, an OpenAI-compatible inference platform. Get an API key at [wallabytoken.com](https://wallabytoken.com) ($0.50 trial credit, no card required): + + + + ```yaml title="config.yaml" + name: My Config + version: 0.0.1 + schema: v1 + + models: + - name: Kimi K3.1 Pro (Wallaby) + provider: openai + model: kimi-k3.1-pro + apiBase: https://api.wallabytoken.com/v1 + apiKey: + ``` + + + +Notes: + +- Use `apiBase` with the full `/v1` suffix; no Responses API flags are needed for this model. +- Kimi K3.1 Pro supports tool use for Agent mode; reasoning tokens are billed as output, matching the official Kimi API. +- Pricing: $1.80 / 1M input ($0.18 cached), $9.00 / 1M output. Machine-readable rate card: [pricing.json](https://wallabytoken.com/pricing.json). Public status page: [status.wallabytoken.com](https://status.wallabytoken.com).