Lumoswitch Docs

Upstream support matrix

Lumoswitch currently includes 33 upstream presets. A preset supplies the protocol, recommended base URL, and model-discovery rules. The console only shows presets available for the current deployment edition and region. Actual models and capabilities follow provider account permissions.

Mainland China and general upstreams

UpstreamProtocolDefault base URL
DeepSeekOpenAI-compatiblehttps://api.deepseek.com/v1
QwenOpenAI-compatiblehttps://dashscope.aliyuncs.com/compatible-mode/v1
GLMOpenAI-compatiblehttps://open.bigmodel.cn/api/paas/v4
KimiOpenAI-compatiblehttps://api.moonshot.cn/v1
DoubaoOpenAI-compatiblehttps://ark.cn-beijing.volces.com/api/v3
HunyuanOpenAI-compatiblehttps://api.hunyuan.cloud.tencent.com/v1
MiniMaxOpenAI-compatiblehttps://api.minimaxi.com/v1
StepFunOpenAI-compatiblehttps://api.stepfun.com/v1
MiMoOpenAI-compatiblehttps://api.xiaomimimo.com/v1
SiliconFlowOpenAI-compatiblehttps://api.siliconflow.cn/v1
QianfanOpenAI-compatiblehttps://qianfan.baidubce.com/v2
BaichuanOpenAI-compatiblehttps://api.baichuan-ai.com/v1
YiOpenAI-compatiblehttps://api.lingyiwanwu.com/v1
ModelScopeOpenAI-compatiblehttps://api-inference.modelscope.cn/v1

International upstreams

UpstreamProtocolDefault base URL
OpenAIOpenAI-compatiblehttps://api.openai.com/v1
OpenRouterOpenAI-compatiblehttps://openrouter.ai/api/v1
AnthropicAnthropichttps://api.anthropic.com
GeminiGeminihttps://generativelanguage.googleapis.com
CohereCoherehttps://api.cohere.com
Mistral AIOpenAI-compatiblehttps://api.mistral.ai/v1
GroqOpenAI-compatiblehttps://api.groq.com/openai/v1
Together AIOpenAI-compatiblehttps://api.together.xyz/v1
Fireworks AIOpenAI-compatiblehttps://api.fireworks.ai/inference/v1
CerebrasOpenAI-compatiblehttps://api.cerebras.ai/v1
NVIDIA NIMOpenAI-compatiblehttps://integrate.api.nvidia.com/v1
PerplexityOpenAI-compatiblehttps://api.perplexity.ai
Novita AIOpenAI-compatiblehttps://api.novita.ai/openai/v1
DeepInfraOpenAI-compatiblehttps://api.deepinfra.com/v1/openai
SambaNovaOpenAI-compatiblehttps://api.sambanova.ai/v1
xAIOpenAI-compatiblehttps://api.x.ai/v1

Local and custom

UpstreamProtocolDefault base URL
OllamaOpenAI-compatiblehttp://localhost:11434/v1
LM StudioOpenAI-compatiblehttp://localhost:1234/v1
Custom OpenAI-compatibleOpenAI-compatibleUser supplied

In a container deployment, localhost points to the Lumoswitch container. Use a host address reachable from the container for Ollama or LM Studio. See Local models.

Verify actual capability

  1. Test the base URL and credential.
  2. Discover models visible to the account.
  3. Send a minimal chat request to the intended model.
  4. Test streaming, tools, vision, reasoning, or embeddings from the target client.

A preset does not grant access to every model or capability. Provider regions, projects, model IDs, and proprietary parameters can differ. OpenAI-compatible does not mean every OpenAI-specific parameter is supported.

On this page