Agent API Documentation

Free, public JSON APIs designed for AI agents. Real-time provider status, model search, pre-flight checks with automatic fallback, trending usage, benchmarks, and infrastructure health. No authentication required.

Base URL: https://aistatus.cc|All endpoints support ?format=text for token-efficient plain text output

Endpoints

GET/api/healthHealth Check

Check if the API is operational and how fresh the data is. Use this to verify the service before making other calls.

Parameters

format"text" for plain text output

Example

curl "https://aistatus.cc/api/health"

Response

{
  "ok": true,
  "dataAge": 3,
  "providersTracked": 12,
  "modelsTracked": 310,
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/providersAll Providers

Returns the current operational status for all 12 tracked AI providers. No UI metadata — clean data for programmatic use.

Parameters

format"text" for plain text output

- status: "operational" | "degraded" | "down" | "unknown"

Example

curl "https://aistatus.cc/api/providers"

Response

{
  "providers": [
    {
      "slug": "anthropic",
      "name": "Anthropic",
      "status": "operational",
      "statusDetail": null,
      "modelCount": 13,
      "statusPageUrl": "https://status.anthropic.com"
    },
    {
      "slug": "openai",
      "name": "OpenAI",
      "status": "degraded",
      "statusDetail": "Partial System Degradation",
      "modelCount": 58,
      "statusPageUrl": "https://status.openai.com"
    }
  ],
  "totalModels": 310,
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/providers/:slugProvider Detail

Returns a single provider's status, all available models with pricing, and recent incidents. One call gives you everything about a provider.

Parameters

slugREQUIREDProvider slug (path parameter)

format"text" for plain text output

Example

curl "https://aistatus.cc/api/providers/anthropic"

Response

{
  "slug": "anthropic",
  "name": "Anthropic",
  "status": "operational",
  "statusDetail": null,
  "modelCount": 13,
  "statusPageUrl": "https://status.anthropic.com",
  "models": [
    {
      "id": "anthropic/claude-opus-4-6",
      "name": "Claude Opus 4.6",
      "context_length": 200000,
      "modality": "text+image->text",
      "pricing": { "prompt": "0.000015", "completion": "0.000075" }
    }
  ],
  "recentIncidents": [
    {
      "from": "operational",
      "to": "degraded",
      "detail": "Elevated error rates",
      "timestamp": "2026-03-11T14:22:00.000Z"
    }
  ],
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/modelsModel Search

Search for models by name or ID, or list all models from a provider. Returns models with their provider's current status inline. At least one of ?q= or ?provider= is required.

Parameters

qModel name or ID to search (fuzzy matching)

providerFilter by provider slug

format"text" for plain text output

- Exact ID match is tried first, then fuzzy name/ID search

- Returns up to 20 results with ?q=, up to 50 with ?provider= only

- Pricing values are in USD per token

Example

curl "https://aistatus.cc/api/models?q=claude-sonnet"

Response

{
  "count": 3,
  "total": 3,
  "models": [
    {
      "id": "anthropic/claude-sonnet-4.6",
      "name": "Claude Sonnet 4.6",
      "provider": {
        "slug": "anthropic",
        "name": "Anthropic",
        "status": "operational",
        "statusDetail": null
      },
      "context_length": 200000,
      "modality": "text+image->text",
      "pricing": { "prompt": "0.000003", "completion": "0.000015" },
      "top_provider": { "context_length": 200000, "max_completion_tokens": 128000 }
    }
  ],
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/models/:provider/:modelSingle Model

Get full detail for a specific model by its ID. Includes provider status, pricing, context length, and modality.

Parameters

providerREQUIREDProvider slug (path parameter)

modelREQUIREDModel name (path parameter)

format"text" for plain text output

Example

curl "https://aistatus.cc/api/models/anthropic/claude-sonnet-4.6"

Response

{
  "id": "anthropic/claude-sonnet-4.6",
  "name": "Claude Sonnet 4.6",
  "provider": {
    "slug": "anthropic",
    "name": "Anthropic",
    "status": "operational",
    "statusDetail": null
  },
  "context_length": 200000,
  "modality": "text+image->text",
  "pricing": { "prompt": "0.000003", "completion": "0.000015" },
  "top_provider": { "context_length": 200000, "max_completion_tokens": 128000 },
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/checkPre-flight Check

The key endpoint for agents. Check if a provider or model is available before making API calls. When the target is down or degraded, returns alternative providers/models automatically.

Parameters

providerProvider slug to check (e.g. anthropic)

modelFull model ID (e.g. anthropic/claude-sonnet-4.6)

format"text" for plain text output

- Use ?provider= OR ?model= (one is required)

- alternatives[] is empty when the target is operational

- Model check also returns pricing, context_length, and modality

Example

curl "https://aistatus.cc/api/check?provider=anthropic"

Response

{
  "provider": "anthropic",
  "name": "Anthropic",
  "status": "degraded",
  "statusDetail": "Elevated error rates",
  "modelCount": 13,
  "topModels": [
    "anthropic/claude-opus-4-6",
    "anthropic/claude-sonnet-4.6"
  ],
  "alternatives": [
    {
      "slug": "openai",
      "name": "OpenAI",
      "status": "operational",
      "suggestedModel": "openai/gpt-4o"
    },
    {
      "slug": "google",
      "name": "Google",
      "status": "operational",
      "suggestedModel": "google/gemini-2.5-pro"
    }
  ],
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/benchmarksLLM Leaderboard

Top 10 models by benchmark score from the HuggingFace Open LLM Leaderboard (average of IFEval, BBH, MATH, GPQA, MUSR, MMLU-PRO).

Parameters

format"text" for plain text output

Example

curl "https://aistatus.cc/api/benchmarks"

Response

{
  "models": [
    {
      "rank": 1,
      "name": "meta-llama/Llama-4-Behemoth-Instruct",
      "avgScore": 72.35,
      "mmluPro": 85.21,
      "mathLvl5": 68.42,
      "gpqa": 62.18,
      "params": 2000,
      "type": "pretrained"
    }
  ],
  "lastUpdated": "2026-03-12T10:00:00.000Z"
}

GET/api/incidentsIncident History

Recent provider status change events. Tracks transitions between operational, degraded, down, and unknown states. Stored persistently in the database.

Parameters

providerFilter by provider slug

format"text" for plain text output

Example

curl "https://aistatus.cc/api/incidents"

Response

{
  "incidents": [
    {
      "slug": "openai",
      "name": "OpenAI",
      "from": "operational",
      "to": "degraded",
      "detail": "Partial System Degradation",
      "timestamp": "2026-03-11T01:35:22.000Z"
    }
  ]
}

GET/api/infra/mcpMCP Infrastructure

Health status of MCP registries, hosted platforms, and official infrastructure with summary counts and recent incidents. Supports ?category= filter (registry, platform, infrastructure).

Parameters

category"registry", "platform", or "infrastructure"

format"text" for plain text output

Example

curl "https://aistatus.cc/api/infra/mcp"

Response

{
  "summary": { "total": 10, "operational": 9, "degraded": 1, "down": 0 },
  "servers": [
    {
      "id": "smithery",
      "name": "Smithery",
      "url": "https://smithery.ai",
      "category": "registry",
      "status": "operational",
      "latencyMs": 245,
      "detail": null,
      "description": "MCP server registry & discovery platform",
      "statusPage": null
    }
  ],
  "recentIncidents": [
    {
      "type": "mcp",
      "serviceId": "composio",
      "name": "Composio",
      "from": "operational",
      "to": "degraded",
      "detail": "Timeout (10s)",
      "timestamp": "2026-03-12T09:15:00.000Z"
    }
  ],
  "lastChecked": "2026-03-12T10:00:00.000Z"
}

GET/api/infra/mcp/:idSingle MCP Service

Detail for a single MCP service by ID, including incident history. Use /api/infra/mcp to see available IDs.

Parameters

format"text" for plain text output

Example

curl "https://aistatus.cc/api/infra/mcp/smithery"

Response

{
  "id": "smithery",
  "name": "Smithery",
  "url": "https://smithery.ai",
  "category": "registry",
  "status": "operational",
  "latencyMs": 245,
  "detail": null,
  "description": "MCP server registry & discovery platform",
  "statusPage": null,
  "recentIncidents": [],
  "lastChecked": "2026-03-12T10:00:00.000Z"
}

GET/api/infra/vectordbVector Infrastructure

Health status of vector databases, platform vector search, and embedding APIs with summary counts and recent incidents. Supports ?category= filter (managed, platform, embedding).

Parameters

category"managed", "platform", or "embedding"

format"text" for plain text output

Example

curl "https://aistatus.cc/api/infra/vectordb"

Response

{
  "summary": { "total": 14, "operational": 13, "degraded": 0, "down": 1 },
  "servers": [
    {
      "id": "pinecone",
      "name": "Pinecone",
      "url": "https://api.pinecone.io/indexes",
      "category": "managed",
      "status": "operational",
      "latencyMs": 180,
      "detail": null,
      "description": "Fully managed serverless vector database",
      "statusPage": "https://status.pinecone.io"
    }
  ],
  "recentIncidents": [],
  "lastChecked": "2026-03-12T10:00:00.000Z"
}

GET/api/infra/vectordb/:idSingle Vector DB Service

Detail for a single Vector DB service by ID, including incident history. Use /api/infra/vectordb to see available IDs.

Parameters

format"text" for plain text output

Example

curl "https://aistatus.cc/api/infra/vectordb/pinecone"

Response

{
  "id": "pinecone",
  "name": "Pinecone",
  "url": "https://api.pinecone.io/indexes",
  "category": "managed",
  "status": "operational",
  "latencyMs": 180,
  "detail": null,
  "description": "Fully managed serverless vector database",
  "statusPage": "https://status.pinecone.io",
  "recentIncidents": [],
  "lastChecked": "2026-03-12T10:00:00.000Z"
}

GET/api/marketModel Market

Browse all AI models from tracked providers with real-time pricing (per 1M tokens), provider status, capabilities, and computed value scores. Supports filtering by provider, capability, budget, and context length.

Parameters

providerFilter by provider slug

capabilityFilter by capability (e.g. "vision")

budgetMax input price per 1M tokens (USD)

min_contextMinimum context length in tokens

sort"value" (default), "price_asc", "price_desc", "context"

limitMax results (default 50, max 200)

format"text" for plain text output

- Pricing is in USD per 1M tokens (not per token)

- valueScore (0-1) combines price efficiency, provider health, context size, and trending popularity

- capabilities are inferred from model modality (e.g. vision for image input support)

Example

curl "https://aistatus.cc/api/market?capability=vision&sort=price_asc&limit=5"

Response

{
  "count": 5,
  "total": 42,
  "models": [
    {
      "id": "google/gemini-2.0-flash-001",
      "name": "Gemini 2.0 Flash",
      "provider": {
        "slug": "google",
        "name": "Google",
        "status": "operational"
      },
      "pricing": {
        "inputPerMillion": 0.1,
        "outputPerMillion": 0.4
      },
      "context_length": 1048576,
      "modality": "text+image->text",
      "capabilities": ["vision"],
      "valueScore": 0.891
    }
  ],
  "lastUpdated": "2026-03-19T10:00:00.000Z"
}

GET/api/market/compareModel Comparison

Side-by-side comparison of 2-10 models. Returns normalized pricing, provider status, context, capabilities, and value scores for each model, sorted by value. Perfect for agents deciding between specific models.

Parameters

modelsREQUIREDComma-separated model IDs (e.g. anthropic/claude-sonnet-4,openai/gpt-4o)

format"text" for plain text output

- Models are sorted by valueScore (highest first)

- Any model IDs not found are returned in a "notFound" array

- Use full model IDs in provider/model format (e.g. anthropic/claude-sonnet-4)

Example

curl "https://aistatus.cc/api/market/compare?models=anthropic/claude-sonnet-4,openai/gpt-4o,google/gemini-2.5-pro"

Response

{
  "models": [
    {
      "id": "google/gemini-2.5-pro",
      "name": "Gemini 2.5 Pro",
      "provider": { "slug": "google", "name": "Google", "status": "operational" },
      "pricing": { "inputPerMillion": 1.25, "outputPerMillion": 10.0 },
      "context_length": 1048576,
      "modality": "text+image->text",
      "capabilities": ["vision"],
      "valueScore": 0.823
    },
    {
      "id": "anthropic/claude-sonnet-4",
      "name": "Claude Sonnet 4",
      "provider": { "slug": "anthropic", "name": "Anthropic", "status": "operational" },
      "pricing": { "inputPerMillion": 3.0, "outputPerMillion": 15.0 },
      "context_length": 200000,
      "modality": "text+image->text",
      "capabilities": ["vision"],
      "valueScore": 0.756
    },
    {
      "id": "openai/gpt-4o",
      "name": "GPT-4o",
      "provider": { "slug": "openai", "name": "OpenAI", "status": "operational" },
      "pricing": { "inputPerMillion": 2.5, "outputPerMillion": 10.0 },
      "context_length": 128000,
      "modality": "text+image->text",
      "capabilities": ["vision"],
      "valueScore": 0.741
    }
  ],
  "lastUpdated": "2026-03-19T10:00:00.000Z"
}

POST/api/hallulens/checkHallucination Check

Verify LLM-generated text against source material. Splits the answer into atomic claims and judges each one against the source using 13 deterministic rules. Returns per-claim verdicts (SUPPORTED, CONTRADICTED, UNSUPPORTED) with evidence, rule IDs, and a groundedness score. No LLM calls — pure rule-based, fully auditable.

Parameters

sourceREQUIREDReference text to verify against (JSON body field, max 50K chars)

answerREQUIREDLLM-generated text to check (JSON body field, max 50K chars)

format"text" for plain text output (query parameter)

Request Body

{
  "source": "The James Webb Space Telescope launched on December 25, 2021, from French Guiana aboard an Ariane 5 rocket. It orbits the Sun at the second Lagrange point (L2), roughly 1.5 million kilometers from Earth. JWST's primary mirror spans 6.5 meters and is composed of 18 hexagonal gold-coated beryllium segments.",
  "answer": "JWST launched in December 2021 from French Guiana. It has a 6.5-meter primary mirror made of 18 hexagonal aluminum segments."
}

- claim.status: "SUPPORTED" | "CONTRADICTED" | "UNSUPPORTED"

- 13 deterministic rules (R-01 through R-13) — no LLM, no network calls

- groundedness = (supported / total) * 100

- verdict: "grounded" (>=90%), "mixed evidence" (>=60%), "hallucinated" (<60%)

- Max 50,000 characters per field

Example

curl -X POST "https://aistatus.cc/api/hallulens/check" \
  -H "Content-Type: application/json" \
  -d '{ "source": "The James Webb Space Telescope launched on December 25, 2021, from French Guiana aboard an Ariane 5 rocket. It orbits the Sun at the second Lagrange point (L2), roughly 1.5 million kilometers from Earth. JWST's primary mirror spans 6.5 meters and is composed of 18 hexagonal gold-coated beryllium segments.", "answer": "JWST launched in December 2021 from French Guiana. It has a 6.5-meter primary mirror made of 18 hexagonal aluminum segments." }'

Response

{
  "tool": "HalluLens",
  "version": "1.1",
  "claims": [
    {
      "text": "JWST launched in December 2021 from French Guiana.",
      "status": "SUPPORTED",
      "rule_id": "R-01",
      "reason": "82% content overlap",
      "evidence": "The James Webb Space Telescope launched on December 25, 2021, from French Guiana aboard an Ariane 5 rocket.",
      "bad_span": ""
    },
    {
      "text": "It has a 6.5-meter primary mirror made of 18 hexagonal aluminum segments.",
      "status": "CONTRADICTED",
      "rule_id": "R-05",
      "reason": "claim says \"aluminum\", source says \"beryllium\"",
      "evidence": "JWST's primary mirror spans 6.5 meters and is composed of 18 hexagonal gold-coated beryllium segments.",
      "bad_span": "aluminum"
    }
  ],
  "summary": {
    "total": 2,
    "supported": 1,
    "contradicted": 1,
    "unsupported": 0,
    "groundedness": 50,
    "verdict": "hallucinated"
  }
}

MCP Integration

Use aistatus.cc as a tool source for MCP-compatible AI agents. The /api/check endpoint is purpose-built for agent pre-flight checks with automatic fallback suggestions.

Use Cases

Pre-flight check: Call /api/check?provider=anthropic before making API calls. Get alternatives if down.

Smart routing: Use /api/check?model=... to verify availability and get fallback models with pricing.

Provider overview: Call /api/providers/:slug for full detail — models, pricing, incidents — in one request.

Market browse: Call /api/market to browse all models with pricing and value scores. Filter by ?capability=vision or ?budget=5.

Model comparison: Use /api/market/compare?models=a,b,c for side-by-side pricing, status, and value analysis.

Smart recommendation: Call /api/market/recommend?task=vision&max_price=5 to get the best model for your task right now.

Infrastructure check: Query /api/infra/mcp and /api/infra/vectordb to verify your RAG pipeline's dependencies.

Hallucination check: POST to /api/hallulens/check with source + answer to verify LLM claims. Get per-claim verdicts and a groundedness score — no LLM needed.

MCP Tool Definition Example

{
  "tools": [
    {
      "name": "check_ai_availability",
      "description": "Pre-flight check: is a provider or model available? Returns alternatives if down/degraded.",
      "endpoint": "https://aistatus.cc/api/check",
      "method": "GET",
      "parameters": {
        "provider": {
          "type": "string",
          "description": "Provider slug to check: anthropic, openai, google, deepseek, meta-llama, mistralai, x-ai, qwen, minimax, stepfun, moonshotai, bytedance-seed",
          "required": false
        },
        "model": {
          "type": "string",
          "description": "Full model ID to check (e.g. anthropic/claude-sonnet-4.6)",
          "required": false
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "get_ai_providers",
      "description": "Get operational status for all AI providers, or detail for a single provider including models and incidents.",
      "endpoint": "https://aistatus.cc/api/providers",
      "method": "GET",
      "parameters": {
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "search_ai_models",
      "description": "Search for AI models by name, or list all models from a provider. Returns availability, pricing, and context length.",
      "endpoint": "https://aistatus.cc/api/models",
      "method": "GET",
      "parameters": {
        "q": {
          "type": "string",
          "description": "Model name or ID to search, e.g. 'claude-sonnet', 'gpt-4o'",
          "required": false
        },
        "provider": {
          "type": "string",
          "description": "Filter by provider slug",
          "required": false
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "get_ai_incidents",
      "description": "Get recent AI provider status change events (outages, degradations, recoveries)",
      "endpoint": "https://aistatus.cc/api/incidents",
      "method": "GET",
      "parameters": {
        "provider": {
          "type": "string",
          "description": "Filter by provider slug",
          "required": false
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "browse_ai_market",
      "description": "Browse AI model market with real-time pricing (per 1M tokens), provider status, and value scores. Filter by provider, capability, budget, context length.",
      "endpoint": "https://aistatus.cc/api/market",
      "method": "GET",
      "parameters": {
        "provider": {
          "type": "string",
          "description": "Filter by provider slug",
          "required": false
        },
        "capability": {
          "type": "string",
          "description": "Filter by capability (e.g. 'vision')",
          "required": false
        },
        "budget": {
          "type": "number",
          "description": "Max input price per 1M tokens in USD",
          "required": false
        },
        "sort": {
          "type": "string",
          "enum": ["value", "price_asc", "price_desc", "context"],
          "required": false
        },
        "limit": {
          "type": "integer",
          "description": "Max results (default 50, max 200)",
          "required": false
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "compare_ai_models",
      "description": "Side-by-side comparison of 2-10 AI models. Returns pricing, status, context, capabilities, and value scores for each model.",
      "endpoint": "https://aistatus.cc/api/market/compare",
      "method": "GET",
      "parameters": {
        "models": {
          "type": "string",
          "description": "Comma-separated model IDs (e.g. 'anthropic/claude-sonnet-4,openai/gpt-4o')",
          "required": true
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "get_ai_recommendation",
      "description": "Get the best AI model recommendations for a task. Only returns operational models, sorted by value score.",
      "endpoint": "https://aistatus.cc/api/market/recommend",
      "method": "GET",
      "parameters": {
        "task": {
          "type": "string",
          "enum": ["general", "vision", "budget", "large_context"],
          "description": "Task type to optimize for",
          "required": false
        },
        "max_price": {
          "type": "number",
          "description": "Max input price per 1M tokens in USD",
          "required": false
        },
        "min_context": {
          "type": "integer",
          "description": "Minimum context length in tokens",
          "required": false
        },
        "provider": {
          "type": "string",
          "description": "Prefer a specific provider slug",
          "required": false
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    },
    {
      "name": "check_hallucination",
      "description": "Verify LLM-generated text against source material. Returns per-claim verdicts (SUPPORTED/CONTRADICTED/UNSUPPORTED) with evidence and a groundedness score. Deterministic, rule-based, no LLM calls.",
      "endpoint": "https://aistatus.cc/api/hallulens/check",
      "method": "POST",
      "parameters": {
        "source": {
          "type": "string",
          "description": "The reference/source text to verify against (max 50K chars)",
          "required": true
        },
        "answer": {
          "type": "string",
          "description": "The LLM-generated text to check for hallucinations (max 50K chars)",
          "required": true
        },
        "format": {
          "type": "string",
          "enum": ["text"],
          "required": false
        }
      }
    }
  ]
}

Agent Workflow Example

Get best model for your task

GET https://aistatus.cc/api/market/recommend?task=vision&max_price=5&format=text

Or compare specific candidates

GET https://aistatus.cc/api/market/compare?models=anthropic/claude-sonnet-4,openai/gpt-4o&format=text

Pre-flight check before API call

GET https://aistatus.cc/api/check?provider=anthropic&format=text

If degraded/down, use suggested alternative

Response includes alternatives[].suggestedModel

Check RAG infra before pipeline run

GET https://aistatus.cc/api/infra/vectordb?category=embedding&format=text

Tracked Providers

anthropicopenaigoogledeepseekmeta-llamamistralaix-aiqwenminimaxstepfunmoonshotaibytedance-seed

OpenAPI Spec

Auto-discovery for agents and tools. The OpenAPI 3.0 specification is available at:

GET https://aistatus.cc/api/openapi.json

Data Sources

Provider Status: OpenRouter Models API + official Atlassian Status Pages (Anthropic, OpenAI, DeepSeek) + Mistral, xAI

Trending Models: OpenRouter Rankings (weekly token volume)

Benchmark Scores: HuggingFace Open LLM Leaderboard

MCP Infrastructure: Direct health checks to registries, platforms, and SDKs

Vector Infrastructure: Direct health checks to managed DBs, platform search, embedding APIs

Model list refreshes every 5s. Status pages, trending, and benchmarks refresh every 60s. MCP and VectorDB refresh every 30s.

◉

Live AI Status Dashboard

Real-time provider status, trending models, and benchmark leaderboards.

View Dashboard →

20:26:01

Saturday, May 30, 2026

aistatus.cc