@@ -278,50 +278,28 @@ messages and normalizes `stats.cached` into `cacheRead`; legacy
278278- Example models: `vercel-ai-gateway/anthropic/claude-opus-4.6`, `vercel-ai-gateway/moonshotai/kimi-k2.6`
279279- CLI: `openclaw onboard --auth-choice ai-gateway-api-key`
280280281-### Kilo Gateway
282-283-- Provider: `kilocode`
284-- Auth: `KILOCODE_API_KEY`
285-- Example model: `kilocode/kilo/auto`
286-- CLI: `openclaw onboard --auth-choice kilocode-api-key`
287-- Base URL: `https://api.kilo.ai/api/gateway/`
288-- Static fallback catalog ships `kilocode/kilo/auto`; live `https://api.kilo.ai/api/gateway/models` discovery can expand the runtime catalog further.
289-- Exact upstream routing behind `kilocode/kilo/auto` is owned by Kilo Gateway, not hard-coded in OpenClaw.
290-291-See [/providers/kilocode](/providers/kilocode) for setup details.
292-293281### Other bundled provider plugins
294282295-| Provider | Id | Auth env | Example model |
296-| --------------------------------------- | -------------------------------- | ------------------------------------------------------------ | ---------------------------------------------------------- |
297-| BytePlus | `byteplus` / `byteplus-plan` | `BYTEPLUS_API_KEY` | `byteplus-plan/ark-code-latest` |
298-| Cerebras | `cerebras` | `CEREBRAS_API_KEY` | `cerebras/zai-glm-4.7` |
299-| Cohere | `cohere` | `COHERE_API_KEY` | `cohere/command-a-03-2025` |
300-| Cloudflare AI Gateway | `cloudflare-ai-gateway` | `CLOUDFLARE_AI_GATEWAY_API_KEY` | - |
301-| DeepInfra | `deepinfra` | `DEEPINFRA_API_KEY` | `deepinfra/deepseek-ai/DeepSeek-V4-Flash` |
302-| DeepSeek | `deepseek` | `DEEPSEEK_API_KEY` | `deepseek/deepseek-v4-flash` |
303-| GitHub Copilot | `github-copilot` | `COPILOT_GITHUB_TOKEN` / `GH_TOKEN` / `GITHUB_TOKEN` | - |
304-| Groq | `groq` | `GROQ_API_KEY` | - |
305-| Hugging Face Inference | `huggingface` | `HUGGINGFACE_HUB_TOKEN` or `HF_TOKEN` | `huggingface/deepseek-ai/DeepSeek-R1` |
306-| Kilo Gateway | `kilocode` | `KILOCODE_API_KEY` | `kilocode/kilo/auto` |
307-| Kimi Coding | `kimi` | `KIMI_API_KEY` or `KIMICODE_API_KEY` | `kimi/kimi-for-coding` |
308-| MiniMax | `minimax` / `minimax-portal` | `MINIMAX_API_KEY` / `MINIMAX_OAUTH_TOKEN` | `minimax/MiniMax-M3` |
309-| Mistral | `mistral` | `MISTRAL_API_KEY` | `mistral/mistral-large-latest` |
310-| Moonshot | `moonshot` | `MOONSHOT_API_KEY` | `moonshot/kimi-k2.6` |
311-| NVIDIA | `nvidia` | `NVIDIA_API_KEY` | `nvidia/nvidia/nemotron-3-ultra-550b-a55b` |
312-| NovitaAI | `novita` | `NOVITA_API_KEY` | `novita/deepseek/deepseek-v3-0324` |
313-| [Ollama Cloud](/providers/ollama-cloud) | `ollama-cloud` | `OLLAMA_API_KEY` | `ollama-cloud/kimi-k2.6` |
314-| OpenRouter | `openrouter` | OpenRouter OAuth or `OPENROUTER_API_KEY` | `openrouter/auto` |
315-| Qianfan | `qianfan` | `QIANFAN_API_KEY` | `qianfan/deepseek-v3.2` |
316-| Qwen Cloud | `qwen` | `QWEN_API_KEY` / `MODELSTUDIO_API_KEY` / `DASHSCOPE_API_KEY` | `qwen/qwen3.5-plus` |
317-| [Qwen OAuth](/providers/qwen-oauth) | `qwen-oauth` | `QWEN_API_KEY` | `qwen-oauth/qwen3.5-plus` |
318-| StepFun | `stepfun` / `stepfun-plan` | `STEPFUN_API_KEY` | `stepfun/step-3.5-flash` |
319-| Together | `together` | `TOGETHER_API_KEY` | `together/meta-llama/Llama-3.3-70B-Instruct-Turbo` |
320-| Venice | `venice` | `VENICE_API_KEY` | - |
321-| Vercel AI Gateway | `vercel-ai-gateway` | `AI_GATEWAY_API_KEY` | `vercel-ai-gateway/anthropic/claude-opus-4.6` |
322-| Volcano Engine (Doubao) | `volcengine` / `volcengine-plan` | `VOLCANO_ENGINE_API_KEY` | `volcengine-plan/ark-code-latest` |
323-| xAI | `xai` | SuperGrok/X Premium OAuth or `XAI_API_KEY` | `xai/grok-4.3` |
324-| Xiaomi | `xiaomi` / `xiaomi-token-plan` | `XIAOMI_API_KEY` / `XIAOMI_TOKEN_PLAN_API_KEY` | `xiaomi/mimo-v2-flash` / `xiaomi-token-plan/mimo-v2.5-pro` |
283+| Provider | Id | Auth env | Example model |
284+| --------------------------------------- | -------------------------------- | ---------------------------------------------------- | ---------------------------------------------------------- |
285+| BytePlus | `byteplus` / `byteplus-plan` | `BYTEPLUS_API_KEY` | `byteplus-plan/ark-code-latest` |
286+| Cohere | `cohere` | `COHERE_API_KEY` | `cohere/command-a-03-2025` |
287+| GitHub Copilot | `github-copilot` | `COPILOT_GITHUB_TOKEN` / `GH_TOKEN` / `GITHUB_TOKEN` | - |
288+| Hugging Face Inference | `huggingface` | `HUGGINGFACE_HUB_TOKEN` or `HF_TOKEN` | `huggingface/deepseek-ai/DeepSeek-R1` |
289+| MiniMax | `minimax` / `minimax-portal` | `MINIMAX_API_KEY` / `MINIMAX_OAUTH_TOKEN` | `minimax/MiniMax-M3` |
290+| Mistral | `mistral` | `MISTRAL_API_KEY` | `mistral/mistral-large-latest` |
291+| Moonshot | `moonshot` | `MOONSHOT_API_KEY` | `moonshot/kimi-k2.6` |
292+| NVIDIA | `nvidia` | `NVIDIA_API_KEY` | `nvidia/nvidia/nemotron-3-ultra-550b-a55b` |
293+| NovitaAI | `novita` | `NOVITA_API_KEY` | `novita/deepseek/deepseek-v3-0324` |
294+| [Ollama Cloud](/providers/ollama-cloud) | `ollama-cloud` | `OLLAMA_API_KEY` | `ollama-cloud/kimi-k2.6` |
295+| OpenRouter | `openrouter` | OpenRouter OAuth or `OPENROUTER_API_KEY` | `openrouter/auto` |
296+| [Qwen OAuth](/providers/qwen-oauth) | `qwen-oauth` | `QWEN_API_KEY` | `qwen-oauth/qwen3.5-plus` |
297+| Together | `together` | `TOGETHER_API_KEY` | `together/meta-llama/Llama-3.3-70B-Instruct-Turbo` |
298+| Venice | `venice` | `VENICE_API_KEY` | - |
299+| Vercel AI Gateway | `vercel-ai-gateway` | `AI_GATEWAY_API_KEY` | `vercel-ai-gateway/anthropic/claude-opus-4.6` |
300+| Volcano Engine (Doubao) | `volcengine` / `volcengine-plan` | `VOLCANO_ENGINE_API_KEY` | `volcengine-plan/ark-code-latest` |
301+| xAI | `xai` | SuperGrok/X Premium OAuth or `XAI_API_KEY` | `xai/grok-4.3` |
302+| Xiaomi | `xiaomi` / `xiaomi-token-plan` | `XIAOMI_API_KEY` / `XIAOMI_TOKEN_PLAN_API_KEY` | `xiaomi/mimo-v2-flash` / `xiaomi-token-plan/mimo-v2.5-pro` |
325303326304#### Quirks worth knowing
327305@@ -341,9 +319,6 @@ See [/providers/kilocode](/providers/kilocode) for setup details.
341319 <Accordion title="xAI">
342320Uses the xAI Responses path. The recommended path is SuperGrok/X Premium OAuth; API keys still work via `XAI_API_KEY` or plugin config, and Grok `web_search` reuses the same auth profile before API-key fallback. `grok-4.3` is the bundled default chat model, and `grok-build-0.1` is selectable for build/coding-focused work. `/fast` or `params.fastMode: true` rewrites `grok-3`, `grok-3-mini`, `grok-4`, and `grok-4-0709` to their `*-fast` variants. `tool_stream` defaults on; disable via `agents.defaults.models["xai/<model>"].params.tool_stream=false`.
343321 </Accordion>
344- <Accordion title="Cerebras">
345-Ships as the bundled `cerebras` provider plugin. GLM uses `zai-glm-4.7`; OpenAI-compatible base URL is `https://api.cerebras.ai/v1`.
346- </Accordion>
347322</AccordionGroup>
348323349324## Providers via `models.providers` (custom/base URL)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。