此 MCP 可以做什么
Provides hosted OpenAI-compatible LLM chat inference, vision inputs, PDF extraction, model discovery, pricing, usage, and optional real-time web search.
工具
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['prompt'], 'properties': {'model': {'type': 'string', 'description': 'Optional model name or alias. Uses default if omitted.'}, 'images': {'type': 'array', 'items': {'type': 'string', 'minLength': 1}, 'description': 'Optional additional images (https URLs or data:image/...;base64,... URIs). Requires a vision-capable model.'}, 'prompt': {'type': 'string', 'minLength': 1, 'description': 'User prompt to send to the selected model.'}, 'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}, 'image_url': {'type': 'string', 'minLength': 1, 'description': 'Optional image as an https URL or data:image/...;base64,... URI. Requires a vision-capable model.'}, 'max_tokens': {'type': 'integer', 'maximum': 4096, 'description': 'Optional max completion tokens.', 'exclusiveMinimum': 0}, 'temperature': {'type': 'number', 'maximum': 2, 'minimum': 0, 'description': 'Optional sampling temperature.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}, 'file_url': {'type': 'string', 'minLength': 1, 'description': 'Optional https URL to fetch the PDF from. Provide this and/or file_base64.'}, 'file_base64': {'type': 'string', 'minLength': 1, 'description': 'Optional base64-encoded PDF bytes (raw base64 or data:application/pdf;base64,...). Provide this and/or file_url.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'properties': {'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['model'], 'properties': {'model': {'type': 'string', 'minLength': 1, 'description': 'Model alias to quote (e.g. llama-3-70b).'}, 'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}, 'max_tokens': {'type': 'integer', 'maximum': 4096, 'description': 'Optional max completion tokens for the quote.', 'exclusiveMinimum': 0}, 'prompt_tokens': {'type': 'integer', 'maximum': 128000, 'description': 'Optional estimated prompt tokens (default: 1).', 'exclusiveMinimum': 0}}, 'additionalProperties': False}
输入模式
{'type': 'object', '$schema': 'http://json-schema.org/draft-07/schema#', 'required': ['query'], 'properties': {'query': {'type': 'string', 'minLength': 1, 'description': 'Search query string.'}, 'api_key': {'type': 'string', 'pattern': '^lmx_[0-9a-f]{32}$', 'description': 'Optional LMX API key (lmx_...). Prefer MCP client Authorization header or env; use this to override per call.'}, 'max_results': {'type': 'integer', 'maximum': 20, 'minimum': 1, 'description': 'Max results to return (1â\x80\x9320, default 5).'}}, 'additionalProperties': False}
近期工具变更
类似的 MCP 服务器
Replicate
Provides access to Replicate models, versions, collections, hardware, predictions, and deployments for running and managing hoste…
deepinfra
Provides access to DeepInfra model and hardware catalogs, inference deployments, deployment metrics, logs, rate limits and usage …
PZERO
Provides prepaid hosted inference for agents, including chat completions, image generation, video generation, model discovery, jo…
fireworks
Inspects and manages Fireworks AI models, deployments, datasets, batch jobs, users, and supervised fine-tuning jobs.
AgentsPodium Hosting
Deploys and manages hosted AI agent instances, including health, billing, LLM keys, peers, webhooks, pausing, and rebuilding.
mcp
Provides access to hosted LLM inference pools, connection details, subscription management, and USDC payments on Base.
AI Compute Radar
Evaluates which tracked AI models fit specific GPUs or Macs and provides model trends, GPU rental prices, and weekly recommendati…
titan
Provides GPU-backed LLM inference on an RTX 4080, with credit purchasing and usage tracking.