proxy.py

from config import Config

from providers.anthropic import HAIKU_5_5_MODEL_ID
from providers.codex_everywhere import CODEX_EVERYWHERE_POOLS, OPENAI_PROTOCOL
from providers.image_relays import image_relay_specs


PROVIDER_DETAILS = {
    'anthropic': {
        'description': 'Anthropic Claude API with Haiku 5.5, native Messages, streaming, tools, and token counting',
        'default_model': HAIKU_5_5_MODEL_ID,
        'endpoints': [
            {
                'url': '/v1/messages',
                'curl': 'curl -X POST "$PROXY_BASE_URL/anthropic/v1/messages" -H "x-api-key: $ADMIN_API_KEY" -H "anthropic-version: 2023-06-01" -H "Content-Type: application/json" -d \'{"model":"claude-haiku-5-5","max_tokens":4096,"messages":[{"role":"user","content":"Hello!"}]}\''
            },
            {
                'url': '/v1/messages/count_tokens',
                'curl': 'curl -X POST "$PROXY_BASE_URL/anthropic/v1/messages/count_tokens" -H "x-api-key: $ADMIN_API_KEY" -H "anthropic-version: 2023-06-01" -H "Content-Type: application/json" -d \'{"model":"claude-haiku-5-5","messages":[{"role":"user","content":"Hello!"}]}\''
            },
            {
                'url': '/v1/models',
                'curl': 'curl "$PROXY_BASE_URL/anthropic/v1/models" -H "x-api-key: $ADMIN_API_KEY"'
            },
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'json_mode': True,
            'json_schema': True,
            'model_discovery': True,
            'raw_passthrough': True,
        },
    },
    'openai': {
        'description': 'OpenAI API for GPT models and embeddings',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/openai/v1/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"gpt-3.5-turbo\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'cerebras': {
        'description': 'Cerebras AI models for text generation and chat',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/cerebras/v1/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"cerebras/btlm-3b-8k\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'googleai': {
        'description': 'Google AI with Gemini models for text and multimodal tasks',
        'endpoints': [
            {
                'url': '/predict',
                'curl': 'curl -X POST "http://localhost:1400/googleai/predict" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"google/gemini-pro\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'xai': {
        'description': 'X.AI (formerly Twitter) language models',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/xai/v1/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"x-1\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'groq': {
        'description': 'Groq API for ultra-fast LLM inference',
        'endpoints': [
            {
                'url': '/openai/v1/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/groq/openai/v1/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"llama3-70b-8192\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'azure': {
        'description': 'Azure AI models for text generation and chat',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/azure/v1/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"gpt-4\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'scaleway': {
        'description': 'Scaleway AI models for text generation',
        'endpoints': [
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/scaleway/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"mistral\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'hyperbolic': {
        'description': 'Hyperbolic AI models for text generation',
        'endpoints': [
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/hyperbolic/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"mixtral-8x7b\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ]
    },
    'sambanova': {
        'description': 'SambaNova AI models for text generation and chat',
        'endpoints': [
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/sambanova/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"sambanova-gpt\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/completions',
                'curl': 'curl -X POST "http://localhost:1400/sambanova/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"sambanova-gpt\\", \\"prompt\\": \\"Hello!\\"}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': False,
            'json_mode': False
        }
    },
    'openrouter': {
        'description': 'OpenRouter - Gateway to multiple AI models including Anthropic, Meta, Google, and more',
        'endpoints': [
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/openrouter/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -H "HTTP-Referer: $YOUR_SITE_URL" -H "X-Title: $YOUR_APP_NAME" -d "{\\"model\\": \\"openai/gpt-3.5-turbo\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/openrouter/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"anthropic/claude-2\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/models',
                'curl': 'curl -X GET "http://localhost:1400/openrouter/models" -H "Authorization: Bearer $API_KEY"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'json_mode': True
        }
    },
    'opencode': {
        'description': 'OpenCode Go subscription - native OpenAI Chat Completions, Responses, Anthropic Messages, streaming, and dynamic model discovery for curated coding models',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/opencode/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"kimi-k3\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}], \\"max_tokens\\": 128}"'
            },
            {
                'url': '/v1/messages',
                'curl': 'curl -X POST "$PROXY_BASE_URL/opencode/v1/messages" -H "X-Api-Key: $ADMIN_API_KEY" -H "Anthropic-Version: 2023-06-01" -H "Content-Type: application/json" -d "{\\"model\\": \\"minimax-m3\\", \\"max_tokens\\": 128, \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/responses',
                'curl': 'curl -X POST "$PROXY_BASE_URL/opencode/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"gpt-5.6-luna\\", \\"input\\": \\"Hello!\\"}"'
            },
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/opencode/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/opencode/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"kimi-k3\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Legacy normalized chat route\\"}]}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'responses': True,
            'anthropic_messages': True,
            'model_discovery': True,
            'json_mode': True
        },
        'default_model': 'kimi-k3'
    },
    'mimo': {
        'description': 'Xiaomi MiMo Token Plan - OpenAI-compatible access to MiMo-V2.5-Pro',
        'endpoints': [
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/mimo/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"mimo-v2.5-pro\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}], \\"max_tokens\\": 128}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'json_mode': True
        },
        'default_model': 'mimo-v2.5-pro'
    },
    'nanogpt': {
        'description': 'NanoGPT raw multi-modal gateway for OpenAI and Anthropic text APIs, media, data, memory, moderation, batches, usage, TEE, and accountless x402 payments',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NANOGPT_MODEL\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/messages',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/messages" -H "X-Api-Key: $ADMIN_API_KEY" -H "Anthropic-Version: 2023-06-01" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NANOGPT_MODEL\\", \\"max_tokens\\": 1024, \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/responses',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NANOGPT_MODEL\\", \\"input\\": \\"Hello!\\"}"'
            },
            {
                'url': '/v1/models?detailed=true',
                'curl': 'curl -X GET "$PROXY_BASE_URL/nanogpt/v1/models?detailed=true" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/images/generations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NANOGPT_IMAGE_MODEL\\", \\"prompt\\": \\"A lighthouse at dusk\\"}"'
            },
            {
                'url': '/generate-video',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/generate-video" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NANOGPT_VIDEO_MODEL\\", \\"prompt\\": \\"A lighthouse at dusk\\"}"'
            },
            {
                'url': '/v1/audio/speech',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/audio/speech" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NANOGPT_TTS_MODEL\\", \\"voice\\": \\"alloy\\", \\"input\\": \\"Welcome to NanoGPT.\\"}"'
            },
            {
                'url': '/v1/audio/transcriptions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/audio/transcriptions" -H "Authorization: Bearer $ADMIN_API_KEY" -F "file=@audio.mp3" -F "model=$NANOGPT_STT_MODEL"'
            },
            {
                'url': '/v1/data/web/search',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/data/web/search" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"query\\": \\"latest AI news\\"}"'
            },
            {
                'url': '/v1/batches',
                'curl': 'curl -X POST "$PROXY_BASE_URL/nanogpt/v1/batches" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"input_file_id\\": \\"file_...\\", \\"endpoint\\": \\"/v1/chat/completions\\", \\"completion_window\\": \\"24h\\"}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True,
            'function_calling': True,
            'json_mode': True,
            'vision': True,
            'embeddings': True,
            'audio': True,
            'images': True,
            'video': True,
            'web_search': True,
            'memory': True,
            'moderation': True,
            'batch': True,
            'x402': True
        }
    },
    'navyai': {
        'description': 'NavyAI raw unified gateway for OpenAI Chat and Responses, Anthropic Messages, images, video jobs, embeddings, speech, moderation, model status, usage, and OAuth tokens',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NAVYAI_MODEL\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/messages',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/messages" -H "X-Api-Key: $ADMIN_API_KEY" -H "Anthropic-Version: 2023-06-01" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NAVYAI_MODEL\\", \\"max_tokens\\": 1024, \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/responses',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NAVYAI_MODEL\\", \\"input\\": \\"Hello!\\"}"'
            },
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/models/status',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/models/status" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/images/generations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NAVYAI_IMAGE_MODEL\\", \\"prompt\\": \\"A naval command room at dusk\\", \\"sync\\": false}"'
            },
            {
                'url': '/v1/images/generations/{id}',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/images/generations/$JOB_ID" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/audio/speech',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/audio/speech" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NAVYAI_TTS_MODEL\\", \\"voice\\": \\"alloy\\", \\"input\\": \\"Welcome aboard.\\"}"'
            },
            {
                'url': '/v1/audio/transcriptions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/audio/transcriptions" -H "Authorization: Bearer $ADMIN_API_KEY" -F "file=@audio.mp3" -F "model=$NAVYAI_STT_MODEL"'
            },
            {
                'url': '/v1/audio/transcriptions/jobs',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/audio/transcriptions/jobs" -H "Authorization: Bearer $ADMIN_API_KEY" -F "file=@audio.mp3" -F "model=$NAVYAI_STT_MODEL"'
            },
            {
                'url': '/v1/audio/transcriptions/jobs/{id}/status',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/audio/transcriptions/jobs/$JOB_ID/status" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/audio/transcriptions/jobs/{id}/download',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/audio/transcriptions/jobs/$JOB_ID/download" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/embeddings',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/embeddings" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$NAVYAI_EMBEDDING_MODEL\\", \\"input\\": \\"NavyAI embeddings\\"}"'
            },
            {
                'url': '/v1/moderations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/moderations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"input\\": \\"Text to review\\"}"'
            },
            {
                'url': '/v1/usage',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/usage" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/oauth/token',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/oauth/token" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"grant_type\\": \\"authorization_code\\", \\"code\\": \\"$NAVY_CODE\\", \\"client_id\\": \\"$NAVY_CLIENT_ID\\", \\"code_verifier\\": \\"$NAVY_CODE_VERIFIER\\"}"'
            },
            {
                'url': '/v1/oauth/me',
                'curl': 'curl -X GET "$PROXY_BASE_URL/navyai/v1/oauth/me" -H "X-MultiLLM-Api-Key: $ADMIN_API_KEY" -H "Authorization: Bearer $NAVY_OAUTH_ACCESS_TOKEN"'
            },
            {
                'url': '/v1/oauth/revoke',
                'curl': 'curl -X POST "$PROXY_BASE_URL/navyai/v1/oauth/revoke" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"token\\": \\"$NAVY_OAUTH_TOKEN\\"}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True,
            'function_calling': True,
            'json_mode': True,
            'vision': True,
            'embeddings': True,
            'audio': True,
            'images': True,
            'video': True,
            'moderation': True,
            'oauth': True,
            'coding_agents': True,
            'roleplay_clients': True
        }
    },
    'codex-easy': {
        'description': 'Codex Everywhere raw OpenAI-compatible gateway; model catalogs are specific to each API-key group',
        'endpoints': [
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/codex-easy/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/responses',
                'curl': 'curl -X POST "$PROXY_BASE_URL/codex-easy/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"grok-4.5\\", \\"reasoning\\": {\\"effort\\": \\"high\\"}, \\"prompt_cache_key\\": \\"conversation-123\\", \\"input\\": \\"Hello!\\", \\"stream\\": true}"'
            },
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/codex-easy/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "X-Grok-Conv-Id: conversation-123" -H "Content-Type: application/json" -d "{\\"model\\": \\"grok-4.5\\", \\"reasoning_effort\\": \\"high\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}], \\"stream\\": true}"'
            },
            {
                'url': '/v1/images/generations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/codex-easy/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$CODEX_EASY_IMAGE_MODEL\\", \\"prompt\\": \\"A small red fox in a forest\\"}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True
        }
    },
    'kimi-code': {
        'description': 'Kimi Code OpenAI-compatible coding endpoint at https://api.kimi.com/coding/v1; Chat Completions model k3',
        'endpoints': [
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/kimi-code/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/kimi-code/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"k3\\", \\"reasoning_effort\\": \\"max\\", \\"prompt_cache_key\\": \\"conversation-123\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}], \\"stream\\": true}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True,
            'function_calling': True
        },
        'default_model': 'k3'
    },
    'cline-pass': {
        'description': 'ClinePass subscription over Cline\'s OpenAI-compatible API at https://api.cline.bot/api/v1; models use full cline-pass/ IDs',
        'endpoints': [
            {
                'url': '/ai/cline/recommended-models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/cline-pass/ai/cline/recommended-models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/cline-pass/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"cline-pass/glm-5.3\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'model_discovery': True,
            'json_mode': True
        },
        'default_model': 'cline-pass/glm-5.3'
    },
    'aihubmix': {
        'description': 'AIHubMix OpenAI-compatible gateway with a seeded free-model catalog and GPT Image, Gemini image, and Doubao Seedream generation',
        'endpoints': [
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/aihubmix/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/aihubmix/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"coding-glm-5.3-free\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/images/generations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/aihubmix/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Idempotency-Key: $REQUEST_ID" -H "Content-Type: application/json" -d "{\\"model\\": \\"gpt-image-2-free\\", \\"prompt\\": \\"A green triangle on white\\", \\"size\\": \\"1024x1024\\", \\"quality\\": \\"low\\", \\"output_format\\": \\"png\\", \\"n\\": 1}"'
            },
            {
                'url': '/v1/images/edits',
                'curl': 'curl -X POST "$PROXY_BASE_URL/aihubmix/v1/images/edits" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Idempotency-Key: $REQUEST_ID" -F "model=gpt-image-2-free" -F "prompt=Change the circle to a square" -F "image=@input.png;type=image/png"'
            },
            {
                'url': '/v1/models/doubao/doubao-seedream-4-0/predictions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/aihubmix/v1/models/doubao/doubao-seedream-4-0/predictions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Idempotency-Key: $REQUEST_ID" -H "Content-Type: application/json" -d "{\\"input\\": {\\"model\\": \\"doubao-seedream-4-0\\", \\"prompt\\": \\"A glass observatory\\", \\"size\\": \\"2K\\", \\"stream\\": false, \\"response_format\\": \\"url\\", \\"watermark\\": false}}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True,
            'images': True,
            'image_editing': True,
            'free_models': True,
            'origin_fallback': True
        },
        'default_model': 'gpt-image-2-free'
    },
    'linkapi': {
        'description': 'LinkAPI multi-cloud gateway with native Claude, Gemini, OpenAI Responses, OpenAI-compatible chat, and image generation/editing',
        'endpoints': [
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "$PROXY_BASE_URL/linkapi/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"'
            },
            {
                'url': '/v1/messages',
                'curl': 'curl -X POST "$PROXY_BASE_URL/linkapi/v1/messages" -H "x-api-key: $ADMIN_API_KEY" -H "anthropic-version: 2023-06-01" -H "Content-Type: application/json" -d "{\\"model\\": \\"$LINKAPI_MODEL\\", \\"max_tokens\\": 1024, \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/responses',
                'curl': 'curl -X POST "$PROXY_BASE_URL/linkapi/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$LINKAPI_MODEL\\", \\"input\\": \\"Hello!\\"}"'
            },
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "$PROXY_BASE_URL/linkapi/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"$LINKAPI_MODEL\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/images/generations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/linkapi/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"gpt-image-2-c\\", \\"prompt\\": \\"A cinematic lighthouse during a storm\\", \\"size\\": \\"1024x1024\\", \\"quality\\": \\"standard\\", \\"style\\": \\"vivid\\", \\"n\\": 1, \\"response_format\\": \\"url\\"}"'
            },
            {
                'url': '/v1/images/edits',
                'curl': 'curl -X POST "$PROXY_BASE_URL/linkapi/v1/images/edits" -H "Authorization: Bearer $ADMIN_API_KEY" -F "model=gpt-image-2-c" -F "prompt=Add soft cinematic lighting" -F "image=@input.png"'
            },
            {
                'url': '/v1beta/models/{model}:generateContent',
                'curl': 'curl -X POST "$PROXY_BASE_URL/linkapi/v1beta/models/gemini-2.5-flash-image:generateContent" -H "x-goog-api-key: $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"contents\\": [{\\"parts\\": [{\\"text\\": \\"Create a watercolor lighthouse at dawn\\"}]}], \\"generationConfig\\": {\\"responseModalities\\": [\\"TEXT\\", \\"IMAGE\\"]}}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True,
            'vision': True,
            'images': True
        }
    },
    'nineteen': {
        'description': 'Nineteen AI - High-performance inference for open-source models',
        'endpoints': [
            {
                'url': '/v1/completions',
                'curl': 'curl -X POST "http://localhost:1400/nineteen/v1/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"TheBloke/Rogue-Rose-103b-v0.2-AWQ\\", \\"prompt\\": \\"Hello!\\", \\"temperature\\": 0.5, \\"max_tokens\\": 50, \\"top_p\\": 0.5, \\"stream\\": true}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': False,
            'json_mode': False
        }
    },
    'palm': {
        'description': 'Google PaLM API for text generation and chat',
        'endpoints': [
            {
                'url': '/models/chat-bison-001/generateText',
                'curl': 'curl -X POST "http://localhost:1400/palm/models/chat-bison-001/generateText" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"prompt\\": {\\"messages\\": [{\\"content\\": \\"Hello!\\"}]}}"'
            },
            {
                'url': '/models/text-bison-001/generateText',
                'curl': 'curl -X POST "http://localhost:1400/palm/models/text-bison-001/generateText" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"prompt\\": \\"Hello!\\"}"'
            },
            {
                'url': '/models',
                'curl': 'curl -X GET "http://localhost:1400/palm/models" -H "Authorization: Bearer $API_KEY"'
            }
        ],
        'supported_features': {
            'streaming': False,
            'function_calling': False,
            'json_mode': False
        },
        'default_model': 'chat-bison-001'
    },
    'together': {
        'description': 'Together AI for high-performance open LLMs and OpenAI-compatible GPT Image 2 generation',
        'endpoints': [
            {
                'url': '/v1/chat/completions',
                'curl': 'curl -X POST "http://localhost:1400/together/v1/chat/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo\\", \\"messages\\": [{\\"role\\": \\"user\\", \\"content\\": \\"Hello!\\"}]}"'
            },
            {
                'url': '/v1/models',
                'curl': 'curl -X GET "http://localhost:1400/together/v1/models" -H "Authorization: Bearer $API_KEY"'
            },
            {
                'url': '/v1/images/generations',
                'curl': 'curl -X POST "$PROXY_BASE_URL/together/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"openai/gpt-image-2\\", \\"prompt\\": \\"A cinematic lighthouse at dusk\\", \\"size\\": \\"1024x1024\\"}"'
            },
            {
                'url': '/v1/completions',
                'curl': 'curl -X POST "http://localhost:1400/together/v1/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo\\", \\"prompt\\": \\"Hello!\\"}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'images': True,
            'json_mode': True
        },
        'default_model': 'meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo'
    },
    'chutes': {
        'description': 'Chutes AI - Access to DeepSeek and other models',
        'endpoints': [
            {
                'url': '/v1/completions',
                'curl': 'curl -X POST "http://localhost:1400/chutes/v1/completions" -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" -d "{\\"model\\": \\"deepseek-ai/DeepSeek-V3\\", \\"prompt\\": \\"Hello!\\", \\"temperature\\": 0.7, \\"max_tokens\\": 100, \\"stream\\": true}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': False,
            'json_mode': False
        },
        'default_model': 'deepseek-ai/DeepSeek-V3'
    },
    'gemini': {
        'description': 'Google Gemini models via Generative Language API',
        'endpoints': [
            {
                'url': '/models/gemini-3.8-flash:generateContent',
                'curl': 'curl -X POST "http://localhost:1400/gemini/models/gemini-3.8-flash:generateContent" -H "Content-Type: application/json" -d "{\\"contents\\": [{\\"parts\\":[{\\"text\\": \\"Explain how AI works\\"}]}], \\"safetySettings\\": [{\\"category\\": \\"HARM_CATEGORY_HARASSMENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_HATE_SPEECH\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_SEXUALLY_EXPLICIT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_DANGEROUS_CONTENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}]}"'
            },
            {
                'url': '/models/gemini-3.5-flash-lite:generateContent',
                'curl': 'curl -X POST "http://localhost:1400/gemini/models/gemini-3.5-flash-lite:generateContent" -H "Content-Type: application/json" -d "{\\"contents\\": [{\\"parts\\":[{\\"text\\": \\"Explain how AI works\\"}]}], \\"safetySettings\\": [{\\"category\\": \\"HARM_CATEGORY_HARASSMENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_HATE_SPEECH\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_SEXUALLY_EXPLICIT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_DANGEROUS_CONTENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}]}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'json_mode': True,
            'web_search': True  # Gemini has web search capability
        },
        'default_model': 'gemini-3.8-flash'
    },
    'gemma': {
        'description': 'Google Gemma open-source models via Generative Language API',
        'endpoints': [
            {
                'url': '/models/gemma-4-26b-a4b-it:generateContent',
                'curl': 'curl -X POST "http://localhost:1400/gemma/models/gemma-4-26b-a4b-it:generateContent" -H "Content-Type: application/json" -d "{\\"contents\\": [{\\"parts\\":[{\\"text\\": \\"Explain how AI works\\"}]}], \\"safetySettings\\": [{\\"category\\": \\"HARM_CATEGORY_HARASSMENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_HATE_SPEECH\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_SEXUALLY_EXPLICIT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_DANGEROUS_CONTENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}]}"'
            },
            {
                'url': '/models/gemma-4-31b-it:generateContent',
                'curl': 'curl -X POST "http://localhost:1400/gemma/models/gemma-4-31b-it:generateContent" -H "Content-Type: application/json" -d "{\\"contents\\": [{\\"parts\\":[{\\"text\\": \\"Explain how AI works\\"}]}], \\"safetySettings\\": [{\\"category\\": \\"HARM_CATEGORY_HARASSMENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_HATE_SPEECH\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_SEXUALLY_EXPLICIT\\", \\"threshold\\": \\"BLOCK_NONE\\"}, {\\"category\\": \\"HARM_CATEGORY_DANGEROUS_CONTENT\\", \\"threshold\\": \\"BLOCK_NONE\\"}]}"'
            }
        ],
        'supported_features': {
            'streaming': True,
            'function_calling': True,
            'json_mode': True
        },
        'default_model': 'gemma-4-26b-a4b-it'
    }
}


def _image_relay_provider_details(spec):
    endpoints = [
        {
            'url': '/v1/models',
            'curl': f'curl -X GET "$PROXY_BASE_URL/{spec.provider}/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"',
        },
        {
            'url': '/v1/images/generations',
            'curl': f'curl -X POST "$PROXY_BASE_URL/{spec.provider}/v1/images/generations" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d \'{{"model":"{spec.models[0]}","prompt":"A cinematic lighthouse at dusk","size":"1024x1024"}}\'',
        },
    ]
    if spec.supports_edits:
        endpoints.append(
            {
                'url': '/v1/images/edits',
                'curl': f'curl -X POST "$PROXY_BASE_URL/{spec.provider}/v1/images/edits" -H "Authorization: Bearer $ADMIN_API_KEY" -F "model={spec.models[0]}" -F "prompt=Add soft cinematic lighting" -F "image=@input.png"',
            }
        )
    if spec.supports_chat:
        endpoints.extend(
            [
                {
                    'url': '/v1/chat/completions',
                    'curl': f'curl -X POST "$PROXY_BASE_URL/{spec.provider}/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d \'{{"model":"{spec.models[0]}","messages":[{{"role":"user","content":"Hello"}}]}}\'',
                },
                {
                    'url': '/v1/responses',
                    'curl': f'curl -X POST "$PROXY_BASE_URL/{spec.provider}/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d \'{{"model":"{spec.models[0]}","input":"Hello"}}\'',
                },
            ]
        )
    return {
        'description': (
            f'{spec.display_name} credential-isolated OpenAI-compatible image relay. '
            'Only model discovery, chat, Responses, and image generation/edit paths are exposed.'
        ),
        'endpoints': endpoints,
        'supported_features': {
            'streaming': spec.supports_chat,
            'images': True,
            'image_edits': spec.supports_edits,
            'model_discovery': True,
            'raw_passthrough': True,
        },
        'default_model': spec.models[0],
    }


def _codex_everywhere_provider_details(pool):
    """One Codex Everywhere pool: its own key, the shared origin, and its group's models."""
    if pool.protocol == OPENAI_PROTOCOL:
        model = "grok-4.7" if pool.provider == "ce-grok-heavy" else "gpt-6.1-sol"
        generation = [
            {
                'url': '/v1/chat/completions',
                'curl': f'curl -X POST "$PROXY_BASE_URL/{pool.provider}/v1/chat/completions" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d \'{{"model":"{model}","messages":[{{"role":"user","content":"Hello"}}]}}\'',
            },
            {
                'url': '/v1/responses',
                'curl': f'curl -X POST "$PROXY_BASE_URL/{pool.provider}/v1/responses" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d \'{{"model":"{model}","input":"Hello"}}\'',
            },
        ]
    else:
        model = "claude-fable-5-1" if pool.provider == "ce-claude-max" else "claude-opus-5-5"
        generation = [
            {
                'url': '/v1/messages',
                'curl': f'curl -X POST "$PROXY_BASE_URL/{pool.provider}/v1/messages" -H "Authorization: Bearer $ADMIN_API_KEY" -H "Content-Type: application/json" -d \'{{"model":"{model}","max_tokens":256,"messages":[{{"role":"user","content":"Hello"}}]}}\'',
            },
        ]
    return {
        'description': (
            f'{pool.display_name}: a Codex Everywhere key group billed at '
            f'{pool.price_multiplier:g}x the vendor list price. The model catalog is specific to the key.'
        ),
        'endpoints': [
            {
                'url': '/v1/models',
                'curl': f'curl -X GET "$PROXY_BASE_URL/{pool.provider}/v1/models" -H "Authorization: Bearer $ADMIN_API_KEY"',
            },
            *generation,
        ],
        'supported_features': {
            'streaming': True,
            'raw_streaming': True,
            'model_discovery': True,
            'raw_passthrough': True,
        },
        'default_model': model,
    }


PROVIDER_DETAILS.update(
    {
        pool.provider: _codex_everywhere_provider_details(pool)
        for pool in CODEX_EVERYWHERE_POOLS
    }
)
PROVIDER_DETAILS.update(
    {
        spec.provider: _image_relay_provider_details(spec)
        for spec in image_relay_specs()
    }
)