{"object":"service_index","name":"llm-gateway","version":"1.0.0","description":"OpenAI-compatible gateway that auto-discovers local model servers (vLLM, llama.cpp, SGLang) and routes inference by model name. Read-only discovery is also mirrored for the vLLM, LM Studio and Ollama dialects.","registry":"https://sevenk.ai/models","models":"https://sevenk.ai/v1/models","well_known":"https://sevenk.ai/.well-known/models","endpoints":{"chat_completions":"https://sevenk.ai/v1/chat/completions","completions":"https://sevenk.ai/v1/completions","embeddings":"https://sevenk.ai/v1/embeddings","responses":"https://sevenk.ai/v1/responses","rerank":"https://sevenk.ai/v1/rerank","audio_transcriptions":"https://sevenk.ai/v1/audio/transcriptions"},"dialects":{"openai":"https://sevenk.ai/v1","vllm":"https://sevenk.ai/vllm/v1","lmstudio":"https://sevenk.ai/lmstudio/v1","ollama":"https://sevenk.ai/ollama/v1"},"auth":{"type":"bearer","required_for":"inference"},"model_count":2}