{
 "site": "https://agentapilist.com",
 "checked": "2026-10-17",
 "currency": "USD",
 "units": {
  "tokens_1m": "USD per 1M input / output tokens",
  "text_units_1k": "USD per 1,000 text units (up to 1,000 characters each)",
  "per_image": "USD per image",
  "requests_1k": "USD per 1,000 calls (each request is one call; tokens do not matter)",
  "free": "Free (no charge per request)"
 },
 "presets": [
  {
   "id": "classification",
   "label": "Classification",
   "input_tokens": 500,
   "output_tokens": 5,
   "text_units": 1,
   "images": 0
  },
  {
   "id": "chatbot",
   "label": "Chatbot turn",
   "input_tokens": 1000,
   "output_tokens": 300,
   "text_units": 1,
   "images": 0
  },
  {
   "id": "agent-loop",
   "label": "Agent loop step",
   "input_tokens": 4000,
   "output_tokens": 500,
   "text_units": 4,
   "images": 0
  }
 ],
 "categories": [
  {
   "id": "all",
   "label": "All categories"
  },
  {
   "id": "decision",
   "label": "Decision models"
  },
  {
   "id": "structured",
   "label": "Structured output"
  },
  {
   "id": "guardrail",
   "label": "Guardrails"
  },
  {
   "id": "classifier",
   "label": "Classifiers"
  },
  {
   "id": "gateway",
   "label": "Gateways"
  },
  {
   "id": "search",
   "label": "Web search & fetch"
  }
 ],
 "volumes": [
  10000,
  100000,
  1000000,
  10000000
 ],
 "formula": {
  "tokens_1m": "requests * (input_tokens * in + output_tokens * out) / 1e6",
  "text_units_1k": "requests * text_units * price / 1000",
  "per_image": "requests * images * price",
  "free": "0"
 },
 "note": "List prices before free allowances and taxes.",
 "prices": [
  {
   "free": null,
   "notes": "Input only; no output, cache-read or cache-write charges. Regional processing premiums and long-context multipliers apply (amounts not stated on the guide).",
   "primary": true,
   "id": "openai-decisions",
   "slug": "openai-decisions",
   "name": "OpenAI Decisions API",
   "plan": "gpt-6-luna",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.1,
   "out": 0,
   "src": "https://developers.openai.com/api/docs/guides/decisions",
   "verified": "2026-10-17",
   "rate": "$0.1 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "OpenRouter listing; the Foundry/Azure price page was not checked.",
   "primary": true,
   "id": "microsoft-decision-1",
   "slug": "microsoft-decision-1",
   "name": "Microsoft-Decision-1",
   "plan": "via OpenRouter (provider Azure)",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.042,
   "out": 0,
   "src": "https://openrouter.ai/microsoft/microsoft-decision-1",
   "verified": "2026-10-17",
   "rate": "$0.042 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Output tokens free; long-context mode costs the same per token.",
   "primary": true,
   "id": "jev-typesafe-ai",
   "slug": "jev-typesafe-ai",
   "name": "Jev (TypeSafe AI)",
   "plan": "jev-1.13 / jev-1.14",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.042,
   "out": 0,
   "src": "https://docs.typesafe.ai/models",
   "verified": "2026-10-17",
   "rate": "$0.042 in / $0 out per 1M tokens"
  },
  {
   "free": "10,000 Neurons/day free on Workers AI (Clef-flash: 3,455 Neurons per 1M input tokens). Not deducted here.",
   "notes": "No output-token charge; media are converted to input tokens.",
   "primary": true,
   "id": "cloudflare-clef-flash",
   "slug": "cloudflare-clef",
   "name": "Cloudflare Clef",
   "plan": "Clef-flash",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.038,
   "out": 0,
   "src": "https://developers.cloudflare.com/workers-ai/platform/pricing/",
   "verified": "2026-10-17",
   "rate": "$0.038 in / $0 out per 1M tokens"
  },
  {
   "free": "10,000 Neurons/day free on Workers AI (Clef-omni: 13,636 Neurons per 1M input tokens). Not deducted here.",
   "notes": "",
   "primary": false,
   "id": "cloudflare-clef-omni",
   "slug": "cloudflare-clef",
   "name": "Cloudflare Clef",
   "plan": "Clef-omni",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.15,
   "out": 0,
   "src": "https://developers.cloudflare.com/workers-ai/platform/pricing/",
   "verified": "2026-10-17",
   "rate": "$0.15 in / $0 out per 1M tokens"
  },
  {
   "free": "10,000 Neurons/day free on Workers AI (Clef: 21,818 Neurons per 1M input tokens). Not deducted here.",
   "notes": "",
   "primary": false,
   "id": "cloudflare-clef",
   "slug": "cloudflare-clef",
   "name": "Cloudflare Clef",
   "plan": "Clef",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.24,
   "out": 0,
   "src": "https://developers.cloudflare.com/workers-ai/platform/pricing/",
   "verified": "2026-10-17",
   "rate": "$0.24 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Output free; no per-request fee. State, questions and images count as input.",
   "primary": true,
   "id": "perplexity-decisions",
   "slug": "perplexity-decisions",
   "name": "Perplexity Decisions API",
   "plan": "pplx-decider-v1.2",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.02,
   "out": 0,
   "src": "https://docs.perplexity.ai/getting-started/pricing",
   "verified": "2026-10-17",
   "rate": "$0.02 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Alias pplx-decider-flash-latest. Output free; no per-request fee.",
   "primary": false,
   "id": "perplexity-decisions-flash",
   "slug": "perplexity-decisions",
   "name": "Perplexity Decisions API",
   "plan": "pplx-decider-flash-v1.2",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.01,
   "out": 0,
   "src": "https://docs.perplexity.ai/getting-started/pricing",
   "verified": "2026-10-17",
   "rate": "$0.01 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Shown as a 50%-off promotion from $0.04 per 1M input tokens; promotion end date not stated. Cached input and output $0.",
   "primary": true,
   "id": "inception-mercury-decide",
   "slug": "inception-mercury-decide",
   "name": "Inception Mercury Decide",
   "plan": "mercury-decide (promo)",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.02,
   "out": 0,
   "src": "https://docs.inceptionlabs.ai/get-started/models",
   "verified": "2026-10-17",
   "rate": "$0.02 in / $0 out per 1M tokens"
  },
  {
   "free": "New orgs get $2 free credit on Liquid Console. Not deducted here.",
   "notes": "",
   "primary": true,
   "id": "liquid-ai-d1",
   "slug": "liquid-ai-d1",
   "name": "Liquid AI d1",
   "plan": "d1",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.02,
   "out": 0,
   "src": "https://www.liquid.ai/pricing",
   "verified": "2026-10-17",
   "rate": "$0.02 in / $0 out per 1M tokens"
  },
  {
   "free": "New orgs get $2 free credit on Liquid Console. Not deducted here.",
   "notes": "Output free; cached input also $0.042. Images 1.5 tokens per 32×32 patch, billed again for each question.",
   "primary": true,
   "id": "liquid-ai-d1-ultra",
   "slug": "liquid-ai-d1-ultra",
   "name": "Liquid AI d1-ultra",
   "plan": "d1-ultra",
   "category": "Decision model",
   "cat": "decision",
   "unit": "tokens_1m",
   "in": 0.042,
   "out": 0,
   "src": "https://www.liquid.ai/pricing",
   "verified": "2026-10-17",
   "rate": "$0.042 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Billed only for HTTP 200 responses with a non-empty result; count does not change the price. Server-side tool calls via DeepSeek Harness, Claude Code or Codex are free until 2026-10-31 16:00 UTC.",
   "primary": true,
   "id": "deepseek-search",
   "slug": "deepseek-search-fetch",
   "name": "DeepSeek Search & Fetch API",
   "plan": "Search",
   "category": "Web search & fetch (agent tool)",
   "cat": "search",
   "unit": "requests_1k",
   "price": 0.75,
   "src": "https://api-docs.deepseek.com/quick_start/pricing",
   "verified": "2026-10-17",
   "rate": "$0.75 per 1,000 calls"
  },
  {
   "free": null,
   "notes": "Billed only when page content is returned.",
   "primary": false,
   "id": "deepseek-fetch",
   "slug": "deepseek-search-fetch",
   "name": "DeepSeek Search & Fetch API",
   "plan": "Fetch",
   "category": "Web search & fetch (agent tool)",
   "cat": "search",
   "unit": "requests_1k",
   "price": 0.75,
   "src": "https://api-docs.deepseek.com/quick_start/pricing",
   "verified": "2026-10-17",
   "rate": "$0.75 per 1,000 calls"
  },
  {
   "free": null,
   "notes": "Only the Jev 1.13 price was checked; other models on the gateway were not.",
   "primary": true,
   "id": "llm-gateway-decisions",
   "slug": "llm-gateway-decisions",
   "name": "LLM Gateway — Decisions",
   "plan": "Jev 1.13 via LLM Gateway",
   "category": "Gateway / proxy",
   "cat": "gateway",
   "unit": "tokens_1m",
   "in": 0.042,
   "out": 0,
   "src": "https://llmgateway.io/changelog/system-one-typed-decisions",
   "verified": "2026-10-17",
   "rate": "$0.042 in / $0 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Charged per enabled filter. Denied topics $0.15/1k, sensitive info $0.10/1k, contextual grounding $0.10/1k, Automated Reasoning $0.17/1k per policy; regex and word filters free. A text unit is up to 1,000 characters.",
   "primary": true,
   "id": "aws-bedrock-guardrails",
   "slug": "aws-bedrock-guardrails",
   "name": "AWS Bedrock Guardrails",
   "plan": "Content filters (text)",
   "category": "Policy guardrail",
   "cat": "guardrail",
   "unit": "text_units_1k",
   "price": 0.15,
   "src": "https://aws.amazon.com/bedrock/pricing/",
   "verified": "2026-10-17",
   "rate": "$0.15 per 1,000 text units"
  },
  {
   "free": null,
   "notes": "Checks-only API, text content only. Prompt attack $0.08/1k, sensitive info $0.10/1k.",
   "primary": false,
   "id": "aws-bedrock-guardrails-checks",
   "slug": "aws-bedrock-guardrails",
   "name": "AWS Bedrock Guardrails",
   "plan": "InvokeGuardrailChecks content filters",
   "category": "Policy guardrail",
   "cat": "guardrail",
   "unit": "text_units_1k",
   "price": 0.07,
   "src": "https://aws.amazon.com/bedrock/pricing/",
   "verified": "2026-10-17",
   "rate": "$0.07 per 1,000 text units"
  },
  {
   "free": null,
   "notes": "",
   "primary": false,
   "id": "aws-bedrock-guardrails-image",
   "slug": "aws-bedrock-guardrails",
   "name": "AWS Bedrock Guardrails",
   "plan": "Content filters (image)",
   "category": "Policy guardrail",
   "cat": "guardrail",
   "unit": "per_image",
   "price": 0.00075,
   "src": "https://aws.amazon.com/bedrock/pricing/",
   "verified": "2026-10-17",
   "rate": "$0.00075 per image"
  },
  {
   "free": "F0 tier: 5,000 text records and 5,000 images per month free; usage stops at the limit. Not deducted here.",
   "notes": "A text record is up to 1,000 characters. Commitment tiers are cheaper at volume.",
   "primary": true,
   "id": "azure-ai-content-safety",
   "slug": "azure-ai-content-safety",
   "name": "Azure AI Content Safety",
   "plan": "S0 text records (East US)",
   "category": "Harm classifier",
   "cat": "classifier",
   "unit": "text_units_1k",
   "price": 0.375,
   "src": "https://prices.azure.com/api/retail/prices",
   "verified": "2026-10-17",
   "rate": "$0.375 per 1,000 text units"
  },
  {
   "free": "F0 tier: 5,000 images per month free. Not deducted here.",
   "notes": "Listed as $0.75 per 1,000 images.",
   "primary": false,
   "id": "azure-ai-content-safety-image",
   "slug": "azure-ai-content-safety",
   "name": "Azure AI Content Safety",
   "plan": "S0 images (East US)",
   "category": "Harm classifier",
   "cat": "classifier",
   "unit": "per_image",
   "price": 0.00075,
   "src": "https://prices.azure.com/api/retail/prices",
   "verified": "2026-10-17",
   "rate": "$0.00075 per image"
  },
  {
   "free": null,
   "notes": "Introductory price through 2026-12-31; standard $1.50 / $7.50 per 1M from 2027-01-01. The response schema counts toward input tokens.",
   "primary": true,
   "id": "google-vertex-structured-output",
   "slug": "google-vertex-structured-output",
   "name": "Gemini structured output (Vertex AI)",
   "plan": "Gemini 3.8 Flash (introductory)",
   "category": "Structured output",
   "cat": "structured",
   "unit": "tokens_1m",
   "in": 0.75,
   "out": 3.75,
   "until": "2026-12-31",
   "after": {
    "in": 1.5,
    "out": 7.5,
    "from": "2027-01-01"
   },
   "src": "https://cloud.google.com/vertex-ai/generative-ai/pricing",
   "verified": "2026-10-17",
   "rate": "$0.75 in / $3.75 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "Price applies to prompts up to 100k tokens. US-only inference_geo costs 1.1x; Batch is 50% off. Structured outputs add an injected system prompt (slightly more input tokens).",
   "primary": true,
   "id": "anthropic-haiku-5-5",
   "slug": "anthropic-claude-structured-output",
   "name": "Anthropic Claude (structured outputs)",
   "plan": "Haiku 5.5",
   "category": "Structured output",
   "cat": "structured",
   "unit": "tokens_1m",
   "in": 0.1,
   "out": 0.5,
   "src": "https://docs.anthropic.com/en/docs/about-claude/pricing",
   "verified": "2026-10-17",
   "rate": "$0.1 in / $0.5 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "",
   "primary": false,
   "id": "anthropic-haiku-4-5",
   "slug": "anthropic-claude-structured-output",
   "name": "Anthropic Claude (structured outputs)",
   "plan": "Haiku 4.5",
   "category": "Structured output",
   "cat": "structured",
   "unit": "tokens_1m",
   "in": 1,
   "out": 5,
   "src": "https://docs.anthropic.com/en/docs/about-claude/pricing",
   "verified": "2026-10-17",
   "rate": "$1 in / $5 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "",
   "primary": false,
   "id": "anthropic-sonnet-5-5",
   "slug": "anthropic-claude-structured-output",
   "name": "Anthropic Claude (structured outputs)",
   "plan": "Sonnet 5.5",
   "category": "Structured output",
   "cat": "structured",
   "unit": "tokens_1m",
   "in": 2,
   "out": 10,
   "src": "https://docs.anthropic.com/en/docs/about-claude/pricing",
   "verified": "2026-10-17",
   "rate": "$2 in / $10 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "",
   "primary": false,
   "id": "anthropic-opus-5-5",
   "slug": "anthropic-claude-structured-output",
   "name": "Anthropic Claude (structured outputs)",
   "plan": "Opus 5.5",
   "category": "Structured output",
   "cat": "structured",
   "unit": "tokens_1m",
   "in": 4,
   "out": 20,
   "src": "https://docs.anthropic.com/en/docs/about-claude/pricing",
   "verified": "2026-10-17",
   "rate": "$4 in / $20 out per 1M tokens"
  },
  {
   "free": null,
   "notes": "The moderation endpoint is free to use.",
   "primary": true,
   "id": "openai-moderation",
   "slug": "openai-moderation",
   "name": "OpenAI Moderation",
   "plan": "omni-moderation-latest",
   "category": "Harm classifier",
   "cat": "classifier",
   "unit": "free",
   "src": "https://platform.openai.com/docs/guides/moderation",
   "verified": "2026-10-17",
   "rate": "Free"
  }
 ],
 "not_comparable": [
  {
   "slug": "openrouter-armor",
   "cat": "guardrail",
   "name": "OpenRouter Armor",
   "reason": "Free during the beta; OpenRouter says GA pricing will be announced later, and your normal model inference is billed separately.",
   "src": "https://openrouter.ai/docs/guides/features/plugins/armor",
   "verified": "2026-10-16"
  },
  {
   "slug": "vercel-ai-gateway-decisions",
   "cat": "gateway",
   "name": "Vercel AI Gateway — Decisions",
   "reason": "Pass-through: no markup, you pay the routed provider’s list price, so the cost depends on which model you route to.",
   "src": "https://vercel.com/docs/ai-gateway/pricing",
   "verified": "2026-10-16"
  },
  {
   "slug": "strands-decider-2b",
   "cat": "decision",
   "name": "Strands Decider 2B",
   "reason": "Open weights; you pay your own compute, which depends on your hardware. No per-request price exists.",
   "src": "https://huggingface.co/",
   "verified": "2026-10-15"
  },
  {
   "slug": "laya",
   "cat": "decision",
   "name": "Laya",
   "reason": "Open weights; you pay your own compute, which depends on your hardware. No per-request price exists.",
   "src": "https://huggingface.co/",
   "verified": "2026-10-15"
  },
  {
   "slug": "llama-cpp-decision-endpoint",
   "cat": "decision",
   "name": "llama.cpp decision endpoint",
   "reason": "Free open-source server; compute cost depends on your hardware.",
   "src": "https://huggingface.co/blog",
   "verified": "2026-10-15"
  },
  {
   "slug": "h2o-lightning-4b",
   "cat": "decision",
   "name": "H2O-Lightning-4B",
   "reason": "Open weights; the $0.021 per 1,000 decisions figure is H2O’s own self-hosting estimate, not a price.",
   "src": "https://huggingface.co/",
   "verified": "2026-10-16"
  },
  {
   "slug": "cohere-classify",
   "cat": "classifier",
   "name": "Cohere Classify",
   "reason": "No Classify price is published on cohere.com/pricing.",
   "src": "https://cohere.com/pricing",
   "verified": "2026-10-16"
  }
 ]
}