{
  "meta": {
    "service": "APIpulse AI Pricing API",
    "version": "1.1.0",
    "updated": "Sep 7, 2026",
    "totalModels": 110,
    "totalMediaModels": 2,
    "totalProviders": 15,
    "license": "CC-BY-4.0",
    "website": "https://getapipulse.com",
    "documentation": "https://getapipulse.com/api-docs.html"
  },
  "providers": [
    {
      "id": "openai",
      "name": "OpenAI",
      "modelCount": 34
    },
    {
      "id": "anthropic",
      "name": "Anthropic",
      "modelCount": 16
    },
    {
      "id": "google",
      "name": "Google",
      "modelCount": 18
    },
    {
      "id": "deepseek",
      "name": "DeepSeek",
      "modelCount": 5
    },
    {
      "id": "mistral",
      "name": "Mistral",
      "modelCount": 11
    },
    {
      "id": "cohere",
      "name": "Cohere",
      "modelCount": 3
    },
    {
      "id": "together",
      "name": "Meta (Together.ai)",
      "modelCount": 5
    },
    {
      "id": "moonshot",
      "name": "Moonshot",
      "modelCount": 3
    },
    {
      "id": "xai",
      "name": "xAI",
      "modelCount": 6
    },
    {
      "id": "ai21",
      "name": "AI21",
      "modelCount": 3
    },
    {
      "id": "qwen",
      "name": "Alibaba/Qwen",
      "modelCount": 2
    },
    {
      "id": "openrouter",
      "name": "InclusionAI (OpenRouter)",
      "modelCount": 1
    },
    {
      "id": "inception",
      "name": "Inception",
      "modelCount": 1
    },
    {
      "id": "ibm",
      "name": "IBM Granite (OpenRouter)",
      "modelCount": 1
    },
    {
      "id": "zai",
      "name": "Z.ai",
      "modelCount": 1
    }
  ],
  "models": [
    {
      "id": "openai-gpt6-astra",
      "name": "GPT-6 Astra",
      "provider": "openai",
      "tier": "Premium",
      "input": 10,
      "output": 50,
      "context": "1.05M",
      "deprecated": false,
      "cachedInput": 1,
      "cacheWrite": 12.5,
      "batchInput": 5,
      "batchOutput": 25,
      "note": "Rolling out through the direct OpenAI API: Trusted Access Program enterprises first, with broader API and ChatGPT access in the coming days. Separately available on OpenRouter as standard, batch, Pro and Pro batch variants at published OpenRouter rates. Do not infer general direct-API availability from OpenRouter. Requests above 272K input use 2x input/cache and 1.5x output rates for the full request.",
      "availability": {
        "directOpenAIApi": false,
        "trustedAccessProgram": true,
        "rolloutStatus": "rolling-out",
        "modelId": "gpt-6-astra",
        "inputTokenLimit": 1050000,
        "maxOutputTokens": 128000,
        "inputModalities": [
          "text",
          "image"
        ],
        "outputModalities": [
          "text"
        ],
        "reasoningLevels": [
          "low",
          "medium",
          "high",
          "xhigh",
          "max"
        ],
        "endpoints": [
          "responses",
          "chat-completions",
          "batch"
        ],
        "computerUse": true,
        "mcp": true,
        "hostedShell": true,
        "webSearch": true,
        "fileSearch": true,
        "codeInterpreter": true,
        "imageGeneration": true,
        "skills": true,
        "applyPatch": true,
        "toolSearch": true,
        "structuredOutputs": true,
        "functionCalling": true,
        "longContextThreshold": 272000,
        "longContextInputMultiplier": 2,
        "longContextOutputMultiplier": 1.5,
        "flexPricing": {
          "input": 5,
          "cachedInput": 0.5,
          "output": 25
        },
        "fastPricing": {
          "input": 20,
          "cachedInput": 2,
          "output": 100,
          "euDataResidency": false
        },
        "openRouter": {
          "verified": "2026-09-07",
          "variants": [
            {
              "modelId": "openai/gpt-6-astra",
              "input": 10,
              "cachedInput": 1,
              "cacheWrite": 12.5,
              "output": 50
            },
            {
              "modelId": "openai/gpt-6-astra:batch",
              "input": 5,
              "cachedInput": 0.5,
              "cacheWrite": 6.25,
              "output": 25
            },
            {
              "modelId": "openai/gpt-6-astra-pro",
              "input": 10,
              "cachedInput": 1,
              "cacheWrite": 12.5,
              "output": 50,
              "reasoningMode": "pro"
            },
            {
              "modelId": "openai/gpt-6-astra-pro:batch",
              "input": 5,
              "cachedInput": 0.5,
              "cacheWrite": 6.25,
              "output": 25,
              "reasoningMode": "pro"
            }
          ]
        }
      }
    },
    {
      "id": "openai-gpt56-sol",
      "name": "GPT-5.6 Sol",
      "provider": "openai",
      "tier": "Premium",
      "input": 4,
      "output": 20,
      "context": "1.05M",
      "deprecated": false,
      "cachedInput": 0.4,
      "listInput": 5,
      "listCachedInput": 0.5,
      "listOutput": 30,
      "promotionStart": "2026-08-21",
      "promotionExpiry": "2026-11-21",
      "promotionEligibility": "Qualifying API usage; included-plan and legacy-credit metering may differ",
      "note": "Temporary promotional API rate available at least through Nov 21, 2026; list price is $5/$0.50/$30 per 1M tokens. Available in Kiro. Sol was not included in the Aug 24 AWS GovCloud announcement for Terra and Luna.",
      "availability": {
        "directOpenAIApi": true,
        "kiro": true,
        "awsBedrockGovCloud": false
      }
    },
    {
      "id": "openai-gpt56-terra",
      "name": "GPT-5.6 Terra",
      "provider": "openai",
      "tier": "Mid",
      "input": 2,
      "output": 12,
      "context": "1.05M",
      "deprecated": false,
      "note": "Also available through Amazon Bedrock in AWS GovCloud (US-East and US-West). Bedrock access and billing are separate from the direct OpenAI API.",
      "availability": {
        "directOpenAIApi": true,
        "kiro": true,
        "awsBedrockGovCloud": true,
        "awsRegions": [
          "us-gov-east-1",
          "us-gov-west-1"
        ]
      }
    },
    {
      "id": "openai-gpt56-luna",
      "name": "GPT-5.6 Luna",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.2,
      "output": 1.2,
      "context": "1.05M",
      "deprecated": false,
      "note": "Also available through Amazon Bedrock in AWS GovCloud (US-East and US-West). Bedrock access and billing are separate from the direct OpenAI API.",
      "availability": {
        "directOpenAIApi": true,
        "kiro": true,
        "awsBedrockGovCloud": true,
        "awsRegions": [
          "us-gov-east-1",
          "us-gov-west-1"
        ]
      }
    },
    {
      "id": "openai-gpt56-cyber",
      "name": "GPT-5.6 Cyber",
      "provider": "openai",
      "tier": "Premium",
      "input": 12.5,
      "output": 75,
      "context": "1.05M",
      "deprecated": false,
      "note": "Daybreak cybersecurity model. Alias: daybreak-red-latest. Specialized for security analysis, threat detection, and vulnerability assessment."
    },
    {
      "id": "openai-gpt55",
      "name": "GPT-5.5",
      "provider": "openai",
      "tier": "Premium",
      "input": 5,
      "output": 30,
      "context": "1.05M",
      "deprecated": false
    },
    {
      "id": "openai-gpt55-pro",
      "name": "GPT-5.5 Pro",
      "provider": "openai",
      "tier": "Premium",
      "input": 30,
      "output": 180,
      "context": "1.05M",
      "deprecated": false
    },
    {
      "id": "openai-gpt55-cyber",
      "name": "GPT-5.5 Cyber",
      "provider": "openai",
      "tier": "Premium",
      "input": 12.5,
      "output": 75,
      "context": "1.05M",
      "deprecated": false,
      "note": "Cybersecurity-focused model in the GPT-5.5 family"
    },
    {
      "id": "openai-gpt53-codex",
      "name": "GPT-5.3 Codex",
      "provider": "openai",
      "tier": "Mid",
      "input": 1.75,
      "output": 14,
      "context": "400K",
      "deprecated": false
    },
    {
      "id": "openai-gpt54",
      "name": "GPT-5.4",
      "provider": "openai",
      "tier": "Mid",
      "input": 2.5,
      "output": 15,
      "context": "1.05M",
      "deprecated": false,
      "note": "Long-context pricing (>272K): 2x input, 1.5x output."
    },
    {
      "id": "openai-gpt54-mini",
      "name": "GPT-5.4 mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.75,
      "output": 4.5,
      "context": "1.05M",
      "deprecated": false,
      "note": "Long-context pricing (>272K): 2x input, 1.5x output."
    },
    {
      "id": "openai-gpt54-nano",
      "name": "GPT-5.4 nano",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.2,
      "output": 1.25,
      "context": "1.05M",
      "deprecated": false,
      "note": "Long-context pricing (>272K): 2x input, 1.5x output."
    },
    {
      "id": "openai-gpt54-pro",
      "name": "GPT-5.4 Pro",
      "provider": "openai",
      "tier": "Premium",
      "input": 30,
      "output": 180,
      "context": "1.05M",
      "deprecated": false,
      "note": "Long-context pricing (>272K): 2x input, 1.5x output."
    },
    {
      "id": "openai-gpt52",
      "name": "GPT-5.2",
      "provider": "openai",
      "tier": "Mid",
      "input": 1.75,
      "output": 14,
      "context": "400K",
      "deprecated": false
    },
    {
      "id": "openai-gpt52-pro",
      "name": "GPT-5.2 Pro",
      "provider": "openai",
      "tier": "Premium",
      "input": 21,
      "output": 168,
      "context": "400K",
      "deprecated": false
    },
    {
      "id": "openai-gpt51",
      "name": "GPT-5.1",
      "provider": "openai",
      "tier": "Mid",
      "input": 1.25,
      "output": 10,
      "context": "272K",
      "deprecated": false
    },
    {
      "id": "openai-gpt5-pro",
      "name": "GPT-5 Pro",
      "provider": "openai",
      "tier": "Premium",
      "input": 15,
      "output": 120,
      "context": "272K",
      "deprecated": false
    },
    {
      "id": "openai-gpt5",
      "name": "GPT-5",
      "provider": "openai",
      "tier": "Premium",
      "input": 1.25,
      "output": 10,
      "context": "272K",
      "deprecated": false,
      "deprecatedDate": "2026-12-11",
      "replacement": "openai-gpt56-sol",
      "note": "Active; shutdown Dec 11 2026, replaced by GPT-5.6 Sol"
    },
    {
      "id": "openai-gpt5-mini",
      "name": "GPT-5 mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.25,
      "output": 2,
      "context": "272K",
      "deprecated": false,
      "deprecatedDate": "2026-12-11",
      "replacement": "openai-gpt56-terra",
      "note": "Active; shutdown Dec 11 2026, replaced by GPT-5.6 Terra"
    },
    {
      "id": "openai-gpt5-nano",
      "name": "GPT-5 nano",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.05,
      "output": 0.4,
      "context": "128K",
      "deprecated": false,
      "note": "Ultra-budget model"
    },
    {
      "id": "openai-gpt-oss-120b",
      "name": "GPT-oss 120B",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.15,
      "output": 0.6,
      "context": "128K",
      "deprecated": false,
      "note": "Open-source model (not available via OpenAI API; self-host or use Hugging Face)"
    },
    {
      "id": "openai-gpt-oss-20b",
      "name": "GPT-oss 20B",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.08,
      "output": 0.35,
      "context": "128K",
      "deprecated": false,
      "note": "Open-source model (not available via OpenAI API; self-host or use Hugging Face)"
    },
    {
      "id": "openai-gpt4o",
      "name": "GPT-4o",
      "provider": "openai",
      "tier": "Mid",
      "input": 2.5,
      "output": 10,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-04-14",
      "replacement": "openai-gpt41",
      "note": "Deprecated Apr 2025; replaced by GPT-4.1 family"
    },
    {
      "id": "openai-gpt4o-mini",
      "name": "GPT-4o mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.15,
      "output": 0.6,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-04-14",
      "replacement": "openai-gpt41-nano",
      "note": "Deprecated Apr 2025; replaced by GPT-4.1 family"
    },
    {
      "id": "openai-gpt41",
      "name": "GPT-4.1",
      "provider": "openai",
      "tier": "Mid",
      "input": 2,
      "output": 8,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "openai-gpt41-mini",
      "name": "GPT-4.1 mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.4,
      "output": 1.6,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "openai-gpt41-nano",
      "name": "GPT-4.1 nano",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.1,
      "output": 0.4,
      "context": "1M",
      "deprecated": false,
      "deprecatedDate": "2026-10-23",
      "replacement": "openai-gpt56-luna",
      "note": "Active; shutdown Oct 23 2026, replaced by GPT-5.6 Luna"
    },
    {
      "id": "openai-o3",
      "name": "o3",
      "provider": "openai",
      "tier": "Mid",
      "input": 2,
      "output": 8,
      "context": "200K",
      "deprecated": false,
      "deprecatedDate": "2026-12-11",
      "replacement": "openai-gpt56-sol",
      "note": "Reasoning model; shutdown Dec 11 2026, replaced by GPT-5.6 Sol"
    },
    {
      "id": "openai-o3-mini",
      "name": "o3-mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 1.1,
      "output": 4.4,
      "context": "200K",
      "deprecated": false,
      "deprecatedDate": "2026-10-23",
      "replacement": "openai-gpt56-sol",
      "note": "Reasoning model; shutdown Oct 23 2026, replaced by GPT-5.6 Sol"
    },
    {
      "id": "openai-o4-mini",
      "name": "o4-mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 1.1,
      "output": 4.4,
      "context": "200K",
      "deprecated": false,
      "deprecatedDate": "2026-10-23",
      "replacement": "openai-gpt56-terra",
      "note": "Reasoning model; shutdown Oct 23 2026, replaced by GPT-5.6 Terra"
    },
    {
      "id": "openai-o3-pro",
      "name": "o3 Pro",
      "provider": "openai",
      "tier": "Premium",
      "input": 20,
      "output": 80,
      "context": "200K",
      "deprecated": false,
      "deprecatedDate": "2026-12-11",
      "replacement": "openai-gpt56-sol",
      "note": "Pro reasoning model; shutdown Dec 11 2026, replaced by GPT-5.6 Sol (pro mode)"
    },
    {
      "id": "openai-o4-mini-deep",
      "name": "o4 Mini Deep Research",
      "provider": "openai",
      "tier": "Budget",
      "input": 1,
      "output": 4,
      "context": "200K",
      "deprecated": true,
      "deprecatedDate": "2026-07-23",
      "replacement": "openai-gpt56-sol",
      "note": "Sunset Jul 23, 2026; replaced by GPT-5.6 Sol"
    },
    {
      "id": "openai-gpt-audio",
      "name": "GPT Audio",
      "provider": "openai",
      "tier": "Mid",
      "input": 2.5,
      "output": 10,
      "context": "128K",
      "deprecated": false,
      "note": "Audio-capable model"
    },
    {
      "id": "openai-gpt-audio-mini",
      "name": "GPT Audio Mini",
      "provider": "openai",
      "tier": "Budget",
      "input": 0.6,
      "output": 2.4,
      "context": "128K",
      "deprecated": false,
      "note": "Audio-capable model"
    },
    {
      "id": "anthropic-opus48",
      "name": "Claude Opus 4.8",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 5,
      "output": 25,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "anthropic-opus47",
      "name": "Claude Opus 4.7",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 5,
      "output": 25,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "anthropic-opus46",
      "name": "Claude Opus 4.6",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 5,
      "output": 25,
      "context": "1M",
      "deprecated": false,
      "note": "Legacy model; consider migrating to Opus 4.8"
    },
    {
      "id": "anthropic-opus45",
      "name": "Claude Opus 4.5",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 5,
      "output": 25,
      "context": "200K",
      "deprecated": false,
      "note": "Legacy model; consider migrating to Opus 4.8"
    },
    {
      "id": "anthropic-opus",
      "name": "Claude Opus 4.1",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 15,
      "output": 75,
      "context": "200K",
      "deprecated": true,
      "deprecatedDate": "2026-08-05",
      "replacement": "anthropic-opus5",
      "note": "Retired Aug 5, 2026; replaced by Claude Opus 5 (per Anthropic migration guide)"
    },
    {
      "id": "anthropic-sonnet5",
      "name": "Claude Sonnet 5",
      "provider": "anthropic",
      "tier": "Mid",
      "input": 2,
      "output": 10,
      "context": "1M",
      "deprecated": false,
      "note": "Standard pricing $2/$10 — confirmed permanent as of Aug 12, 2026 (planned Sep 1 increase to $3/$15 was cancelled)"
    },
    {
      "id": "anthropic-sonnet46",
      "name": "Claude Sonnet 4.6",
      "provider": "anthropic",
      "tier": "Mid",
      "input": 3,
      "output": 15,
      "context": "1M",
      "deprecated": false,
      "note": "Legacy model; consider migrating to Sonnet 5"
    },
    {
      "id": "anthropic-sonnet45",
      "name": "Claude Sonnet 4.5",
      "provider": "anthropic",
      "tier": "Mid",
      "input": 3,
      "output": 15,
      "context": "200K",
      "deprecated": false,
      "note": "Legacy model; consider migrating to Sonnet 5"
    },
    {
      "id": "anthropic-sonnet",
      "name": "Claude Sonnet 4",
      "provider": "anthropic",
      "tier": "Mid",
      "input": 3,
      "output": 15,
      "context": "200K",
      "deprecated": true,
      "deprecatedDate": "2026-08-05",
      "replacement": "anthropic-sonnet46",
      "note": "Retired except on Bedrock and Google Cloud; $3/$15 per Anthropic pricing page"
    },
    {
      "id": "anthropic-haiku",
      "name": "Claude Haiku 4.5",
      "provider": "anthropic",
      "tier": "Budget",
      "input": 1,
      "output": 5,
      "context": "200K",
      "deprecated": false
    },
    {
      "id": "anthropic-opus5",
      "name": "Claude Opus 5",
      "provider": "anthropic",
      "tier": "Active",
      "input": 5,
      "output": 25,
      "context": "1M",
      "deprecated": false,
      "note": "Anthropic recommended default model for complex agentic coding and enterprise work"
    },
    {
      "id": "anthropic-fable5",
      "name": "Claude Fable 5",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 10,
      "output": 50,
      "context": "1M",
      "deprecated": false,
      "note": "Legacy model; replaced by Claude Fable 5.1"
    },
    {
      "id": "anthropic-fable51",
      "name": "Claude Fable 5.1",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 10,
      "output": 50,
      "context": "1M",
      "deprecated": false,
      "cachedInput": 0.25,
      "batchInput": 5,
      "batchOutput": 25,
      "note": "Active; 128K maximum output, text and image input, adaptive thinking. Batch is a pricing/request mode, not a separate model. US-only inference costs 1.1x. Mythos 5.1 uses the same model but is restricted to vetted US organizations and is not a public API alias.",
      "availability": {
        "directAnthropicApi": true,
        "modelId": "claude-fable-5-1",
        "openRouter": true,
        "openRouterModelId": "anthropic/claude-fable-5.1",
        "awsBedrock": true,
        "googleCloud": true,
        "microsoftFoundry": true,
        "batchApi": [
          "anthropic",
          "claude-platform-aws"
        ]
      }
    },
    {
      "id": "anthropic-mythos5",
      "name": "Claude Mythos 5",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 10,
      "output": 50,
      "context": "1M",
      "deprecated": false,
      "note": "Invitation-only via Project Glasswing — not generally available via standard API"
    },
    {
      "id": "anthropic-opus48-fast",
      "name": "Claude Opus 4.8 Fast",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 10,
      "output": 50,
      "context": "1M",
      "deprecated": false,
      "note": "Fast variant of Opus 4.8"
    },
    {
      "id": "anthropic-opus47-fast",
      "name": "Claude Opus 4.7 Fast",
      "provider": "anthropic",
      "tier": "Premium",
      "input": 30,
      "output": 150,
      "context": "1M",
      "deprecated": true,
      "deprecatedDate": "2026-07-24",
      "replacement": "anthropic-opus48-fast",
      "note": "Removed Jul 24, 2026; replaced by Claude Opus 4.8 Fast"
    },
    {
      "id": "google-gemini38-flash",
      "name": "Gemini 3.8 Flash",
      "provider": "google",
      "tier": "Budget",
      "input": 0.75,
      "output": 3.75,
      "context": "1M",
      "deprecated": false,
      "cachedInput": 0.075,
      "batchInput": 0.375,
      "batchOutput": 1.875,
      "listInput": 1.5,
      "listCachedInput": 0.15,
      "listOutput": 7.5,
      "promotionStart": "2026-09-02",
      "promotionExpiry": "2026-12-31",
      "promotionEligibility": "Google Gemini API introductory pricing through Dec 31, 2026; scheduled standard pricing begins Jan 1, 2027",
      "note": "GA production model for long-horizon software engineering and autonomous agents. Introductory standard, cached, Batch, Flex and Priority rates end Dec 31, 2026. Standard prices double Jan 1, 2027. Output pricing includes thinking tokens; computer use remains Preview.",
      "availability": {
        "geminiApi": true,
        "googleAIStudio": true,
        "modelId": "gemini-3.8-flash",
        "status": "GA",
        "releaseDate": "2026-09-02",
        "inputTokenLimit": 1048576,
        "maxOutputTokens": 65536,
        "inputModalities": [
          "text",
          "image",
          "video",
          "audio",
          "pdf"
        ],
        "outputModalities": [
          "text"
        ],
        "thinkingLevels": [
          "low",
          "medium",
          "high"
        ],
        "defaultThinkingLevel": "medium",
        "minimalThinkingSupported": false,
        "functionCalling": true,
        "structuredOutputs": true,
        "computerUse": "preview",
        "batchApi": true,
        "flexInference": {
          "input": 0.375,
          "output": 1.875,
          "futureInput": 0.75,
          "futureOutput": 3.75,
          "effectiveDate": "2027-01-01"
        },
        "priorityInference": {
          "input": 1.35,
          "output": 6.75,
          "futureInput": 2.7,
          "futureOutput": 13.5,
          "effectiveDate": "2027-01-01"
        },
        "futureBatchPricing": {
          "input": 0.75,
          "output": 3.75,
          "effectiveDate": "2027-01-01"
        },
        "antigravityManagedAgentsDefault": true
      }
    },
    {
      "id": "google-gemini37-flash",
      "name": "Gemini 3.7 Flash",
      "provider": "google",
      "tier": "Budget",
      "input": 0.75,
      "output": 3.75,
      "context": "1M",
      "deprecated": false,
      "note": "Promotional pricing through Dec 31, 2026 ($0.75/$3.75); standard price $1.50/$7.50 after. Free tier available."
    },
    {
      "id": "google-gemini36-flash",
      "name": "Gemini 3.6 Flash",
      "provider": "google",
      "tier": "Budget",
      "input": 0.75,
      "output": 3.75,
      "context": "1M",
      "deprecated": false,
      "note": "Promotional pricing through Dec 31, 2026 ($0.75/$3.75); standard price $1.50/$7.50 after. Previous generation."
    },
    {
      "id": "google-gemini35-flash",
      "name": "Gemini 3.5 Flash",
      "provider": "google",
      "tier": "Mid",
      "input": 1.5,
      "output": 9,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "google-gemini35-flash-lite",
      "name": "Gemini 3.5 Flash-Lite",
      "provider": "google",
      "tier": "Budget",
      "input": 0.3,
      "output": 2.5,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "google-gemini-omni-11-flash",
      "name": "Gemini Omni 1.1 Flash",
      "provider": "google",
      "tier": "Mid",
      "input": 1.5,
      "output": 17.5,
      "context": "1.05M",
      "deprecated": false,
      "note": "Video-generation and editing model. Output field records the $17.50/M video-output-token rate; text output is $9/M. Google estimates 720p video at about $0.10/second. Supports 3-10s video, extension, interpolation, and 360p/720p/1080p/4K output; 1080p and 4K are upscaled.",
      "availability": {
        "geminiApi": true,
        "paidTierOnly": true,
        "modelId": "gemini-omni-1.1-flash",
        "previewDeprecationDate": "2026-09-30"
      }
    },
    {
      "id": "google-gemini35-transcribe",
      "name": "Gemini 3.5 Transcribe",
      "provider": "google",
      "tier": "Budget",
      "input": 2,
      "output": 12,
      "context": "Audio",
      "deprecated": false,
      "note": "Recorded speech-to-text. Audio input $2/M tokens and text output $12/M; Google estimates about $0.005/min blended. Supports 85+ languages, diarization, word timestamps, and vocabulary biasing.",
      "availability": {
        "geminiApi": true,
        "modelId": "gemini-3.5-transcribe"
      }
    },
    {
      "id": "google-gemini35-transcribe-live",
      "name": "Gemini 3.5 Transcribe Live",
      "provider": "google",
      "tier": "Budget",
      "input": 3.5,
      "output": 21,
      "context": "10 min",
      "deprecated": false,
      "note": "Live WebSocket speech-to-text. Audio input $3.50/M tokens and text output $21/M; Google estimates about $0.009/min blended. Ten-minute sessions; no speaker diarization or word-level timestamps.",
      "availability": {
        "geminiApi": true,
        "liveApi": true,
        "modelId": "gemini-3.5-transcribe-live"
      }
    },
    {
      "id": "google-gemini31-flash-lite",
      "name": "Gemini 3.1 Flash-Lite",
      "provider": "google",
      "tier": "Budget",
      "input": 0.25,
      "output": 1.5,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "google-gemini3-pro",
      "name": "Gemini 3.1 Pro",
      "provider": "google",
      "tier": "Mid",
      "input": 2,
      "output": 12,
      "context": "1M",
      "deprecated": false,
      "note": "Still in preview (gemini-3.1-pro-preview). $2/$12 for ≤200K context; $4/$18 for >200K context."
    },
    {
      "id": "google-gemini3-flash",
      "name": "Gemini 3 Flash",
      "provider": "google",
      "tier": "Budget",
      "input": 0.5,
      "output": 3,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "google-pro",
      "name": "Gemini 2.5 Pro",
      "provider": "google",
      "tier": "Mid",
      "input": 1.25,
      "output": 10,
      "context": "1M",
      "deprecated": false,
      "note": "$1.25/$10 for ≤200K context; $2.50/$15 for >200K context"
    },
    {
      "id": "google-25-flash-lite",
      "name": "Gemini 2.5 Flash-Lite",
      "provider": "google",
      "tier": "Budget",
      "input": 0.1,
      "output": 0.4,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "google-25-flash",
      "name": "Gemini 2.5 Flash",
      "provider": "google",
      "tier": "Mid",
      "input": 0.3,
      "output": 2.5,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "google-flash",
      "name": "Gemini 2.0 Flash",
      "provider": "google",
      "tier": "Budget",
      "input": 0.1,
      "output": 0.4,
      "context": "1M",
      "deprecated": true,
      "deprecatedDate": "2026-06-01",
      "replacement": "google-gemini3-flash",
      "note": "Shut down Jun 1, 2026; replaced by Gemini 3 Flash"
    },
    {
      "id": "google-flash-lite",
      "name": "Gemini 2.0 Flash Lite",
      "provider": "google",
      "tier": "Budget",
      "input": 0.075,
      "output": 0.3,
      "context": "1M",
      "deprecated": true,
      "deprecatedDate": "2026-06-01",
      "replacement": "google-gemini31-flash-lite",
      "note": "Shut down Jun 1, 2026; replaced by Gemini 3.1 Flash-Lite"
    },
    {
      "id": "google-nano-banana2",
      "name": "Nano Banana 2 (Gemini 3.1 Flash Image)",
      "provider": "google",
      "tier": "Budget",
      "input": 0.5,
      "output": 3,
      "context": "131K",
      "deprecated": false,
      "note": "Image generation model based on Gemini 3.1 Flash"
    },
    {
      "id": "google-nano-banana-pro",
      "name": "Nano Banana Pro (Gemini 3 Pro Image)",
      "provider": "google",
      "tier": "Mid",
      "input": 2,
      "output": 12,
      "context": "65K",
      "deprecated": false,
      "note": "Image generation model based on Gemini 3 Pro"
    },
    {
      "id": "deepseek-v4-pro",
      "name": "DeepSeek V4 Pro",
      "provider": "deepseek",
      "tier": "Budget",
      "input": 0.66,
      "output": 1.98,
      "context": "1M",
      "deprecated": false,
      "note": "Peak/off-peak pricing since Aug 16 2026. Off-peak: $0.66/$1.98; Peak (01:00-04:00, 06:00-10:00 UTC): $1.32/$3.96. Cached input: $0.022 (off-peak) / $0.044 (peak). Since Aug 23: weekends (Sat/Sun Beijing time) are all-day off-peak."
    },
    {
      "id": "deepseek-v4-flash",
      "name": "DeepSeek V4 Flash",
      "provider": "deepseek",
      "tier": "Budget",
      "input": 0.22,
      "output": 0.66,
      "context": "1M",
      "deprecated": false,
      "note": "Peak/off-peak pricing since Aug 16 2026. Off-peak: $0.22/$0.66; Peak (01:00-04:00, 06:00-10:00 UTC): $0.44/$1.32. Cached input: $0.007 (off-peak) / $0.014 (peak). Since Aug 23: weekends (Sat/Sun Beijing time) are all-day off-peak."
    },
    {
      "id": "deepseek-v4-flash-vision-exp",
      "name": "DeepSeek V4 Flash Vision Exp",
      "provider": "deepseek",
      "tier": "Budget",
      "input": 0.22,
      "output": 0.66,
      "context": "1M",
      "deprecated": false,
      "note": "Vision-capable experimental variant of V4 Flash. Same pricing. Peak/off-peak: same as V4 Flash. Images converted to tokens based on dimensions."
    },
    {
      "id": "deepseek-v32",
      "name": "DeepSeek V3.2",
      "provider": "deepseek",
      "tier": "Budget",
      "input": 0.23,
      "output": 0.34,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-07-07",
      "replacement": "deepseek-v4-flash",
      "note": "No longer listed on DeepSeek pricing page; V4 Flash is the successor"
    },
    {
      "id": "deepseek-v3",
      "name": "DeepSeek V3",
      "provider": "deepseek",
      "tier": "Budget",
      "input": 0.27,
      "output": 1.1,
      "context": "128K",
      "deprecated": true,
      "replacement": "deepseek-v4-flash"
    },
    {
      "id": "mistral-large",
      "name": "Mistral Large 3",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.5,
      "output": 1.5,
      "context": "262K",
      "deprecated": false
    },
    {
      "id": "mistral-medium",
      "name": "Mistral Medium 3.5",
      "provider": "mistral",
      "tier": "Mid",
      "input": 1.5,
      "output": 7.5,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "mistral-small",
      "name": "Mistral Small 4",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.15,
      "output": 0.6,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "codestral",
      "name": "Codestral",
      "provider": "mistral",
      "tier": "Mid",
      "input": 0.3,
      "output": 0.9,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "magistral-medium",
      "name": "Magistral Medium",
      "provider": "mistral",
      "tier": "Mid",
      "input": 2,
      "output": 5,
      "context": "256K",
      "deprecated": true,
      "deprecatedDate": "2026-07-31",
      "replacement": "mistral-medium",
      "note": "Retired Jul 31 2026; replaced by Mistral Medium 3.5"
    },
    {
      "id": "magistral-small",
      "name": "Magistral Small",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.5,
      "output": 1.5,
      "context": "256K",
      "deprecated": true,
      "deprecatedDate": "2026-07-31",
      "replacement": "mistral-small",
      "note": "Retired Jul 31 2026; replaced by Mistral Small 4"
    },
    {
      "id": "devstral2",
      "name": "Devstral 2",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.4,
      "output": 2,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-07-31",
      "replacement": "mistral-medium",
      "note": "Retired Jul 31 2026; replaced by Mistral Medium 3.5"
    },
    {
      "id": "devstral-small2",
      "name": "Devstral Small 2",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.1,
      "output": 0.3,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-03-31",
      "replacement": "mistral-medium",
      "note": "Retired Mar 31 2026; replaced by Mistral Medium 3.5"
    },
    {
      "id": "ministral-3-3b",
      "name": "Ministral 3 3B",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.1,
      "output": 0.1,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "ministral-3-8b",
      "name": "Ministral 3 8B",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.15,
      "output": 0.15,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "ministral-3-14b",
      "name": "Ministral 3 14B",
      "provider": "mistral",
      "tier": "Budget",
      "input": 0.2,
      "output": 0.2,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "cohere-command-a",
      "name": "Command A",
      "provider": "cohere",
      "tier": "Mid",
      "input": 2.5,
      "output": 10,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "cohere-command-r-plus",
      "name": "Command R+",
      "provider": "cohere",
      "tier": "Mid",
      "input": 2.5,
      "output": 10,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "cohere-command-r",
      "name": "Command R",
      "provider": "cohere",
      "tier": "Budget",
      "input": 0.5,
      "output": 1.5,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "llama-4-scout",
      "name": "Llama 4 Scout",
      "provider": "together",
      "tier": "Budget",
      "input": 0.18,
      "output": 0.59,
      "context": "1M",
      "deprecated": false,
      "note": "Delisted from Together.ai serverless Jul 2026 — may only be available via dedicated endpoints"
    },
    {
      "id": "llama-4-maverick",
      "name": "Llama 4 Maverick",
      "provider": "together",
      "tier": "Budget",
      "input": 0.27,
      "output": 0.85,
      "context": "1M",
      "deprecated": false,
      "note": "Delisted from Together.ai serverless Jul 2026 — may only be available via dedicated endpoints"
    },
    {
      "id": "llama-33-70b",
      "name": "Llama 3.3 70B",
      "provider": "together",
      "tier": "Mid",
      "input": 1.04,
      "output": 1.04,
      "context": "128K",
      "deprecated": false
    },
    {
      "id": "llama-3.1-70b",
      "name": "Llama 3.1 70B",
      "provider": "together",
      "tier": "Mid",
      "input": 0.88,
      "output": 0.88,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-07-01",
      "replacement": "llama-33-70b"
    },
    {
      "id": "llama-3.1-8b",
      "name": "Llama 3.1 8B",
      "provider": "together",
      "tier": "Budget",
      "input": 0.14,
      "output": 0.14,
      "context": "128K",
      "deprecated": true,
      "deprecatedDate": "2026-07-01",
      "replacement": "llama-33-70b"
    },
    {
      "id": "kimi-k3",
      "name": "Kimi K3",
      "provider": "moonshot",
      "tier": "Mid",
      "input": 2.9,
      "output": 14,
      "context": "1M",
      "deprecated": false,
      "note": "Moonshot's flagship model for long-form coding and knowledge work, always-on reasoning"
    },
    {
      "id": "kimi-k26",
      "name": "Kimi K2.6",
      "provider": "moonshot",
      "tier": "Budget",
      "input": 0.95,
      "output": 4,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "kimi-k27-code",
      "name": "Kimi K2.7 Code",
      "provider": "moonshot",
      "tier": "Budget",
      "input": 0.95,
      "output": 4,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "xai-grok3",
      "name": "Grok 4.3",
      "provider": "xai",
      "tier": "Mid",
      "input": 1.25,
      "output": 2.5,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "xai-grok3-mini",
      "name": "Grok Build 0.1",
      "provider": "xai",
      "tier": "Budget",
      "input": 1,
      "output": 2,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "xai-grok45",
      "name": "Grok 4.5",
      "provider": "xai",
      "tier": "Mid",
      "input": 2,
      "output": 6,
      "context": "500K",
      "deprecated": false,
      "note": "xAI's recommended model for most use cases"
    },
    {
      "id": "xai-grok420",
      "name": "Grok 4.20",
      "provider": "xai",
      "tier": "Mid",
      "input": 1.25,
      "output": 2.5,
      "context": "1M",
      "deprecated": false
    },
    {
      "id": "xai-grok420-multi",
      "name": "Grok 4.20 Multi-Agent",
      "provider": "xai",
      "tier": "Mid",
      "input": 1.25,
      "output": 2.5,
      "context": "1M",
      "deprecated": false,
      "note": "Multi-agent variant of Grok 4.20 (grok-4.20-multi-agent-0309). Same pricing as Grok 4.20: $1.25/$2.50 for <200K context; $2.50/$5.00 for ≥200K."
    },
    {
      "id": "xai-grok46",
      "name": "Grok 4.6",
      "provider": "xai",
      "tier": "Mid",
      "input": 2,
      "output": 6,
      "context": "500K",
      "deprecated": false,
      "note": "xAI's newest model (default for code and chat). Tiered pricing: $2/$6 for <200K context; $4/$12 for ≥200K."
    },
    {
      "id": "ai21-jamba-mini",
      "name": "Jamba Mini",
      "provider": "ai21",
      "tier": "Budget",
      "input": 0.2,
      "output": 0.4,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "ai21-jamba17",
      "name": "Jamba 1.7 Large",
      "provider": "ai21",
      "tier": "Mid",
      "input": 2,
      "output": 8,
      "context": "256K",
      "deprecated": false
    },
    {
      "id": "ai21-jamba",
      "name": "Jamba 1.5 Large",
      "provider": "ai21",
      "tier": "Mid",
      "input": 2,
      "output": 8,
      "context": "256K",
      "deprecated": true,
      "replacement": "ai21-jamba17"
    },
    {
      "id": "qwen-qwen37-flash",
      "name": "Qwen 3.7 Flash",
      "provider": "qwen",
      "tier": "Budget",
      "input": 0.03,
      "output": 0.13,
      "context": "1M",
      "deprecated": false,
      "note": "Cheapest multimodal model available — text+image input, text output"
    },
    {
      "id": "qwen-qwen38-flash",
      "name": "Qwen 3.8 Flash",
      "provider": "qwen",
      "tier": "Budget",
      "input": 0.16,
      "output": 0.47,
      "context": "1M",
      "deprecated": false,
      "note": "OpenRouter model qwen/qwen3.8-flash; released Aug 26, 2026; tool calling and JSON-schema structured outputs. Hosted Flash is distinct from the open-weight experimental Qwen3.8-Flash-Next model."
    },
    {
      "id": "openrouter-ling30-flash-fin-free",
      "name": "Ling 3.0 Flash Fin (Free)",
      "provider": "openrouter",
      "tier": "Free",
      "input": 0,
      "output": 0,
      "context": "262K",
      "deprecated": false,
      "note": "Finance-focused 124B-total/5.1B-active MoE. Free, rate-limited OpenRouter endpoint; 32,768 max output tokens and tool calling. response_format JSON enforcement is not supported. Validate outputs and do not treat model responses as financial advice.",
      "availability": {
        "openRouter": true,
        "modelId": "inclusionai/ling-3.0-flash-fin:free",
        "rateLimited": true
      }
    },
    {
      "id": "inception-mercury25-preview",
      "name": "Mercury 2.5 Preview",
      "provider": "inception",
      "tier": "Budget",
      "input": 0.2,
      "output": 0.75,
      "context": "260K",
      "deprecated": false,
      "cachedInput": 0.02,
      "note": "Preview/early-access diffusion language model with reasoning, tool use and structured output. Canonical prices are Inception direct rates; OpenRouter's launch prices are a temporary provider promotion and must not be treated as universal list pricing.",
      "availability": {
        "directInceptionApi": true,
        "modelId": "mercury-2.5-preview",
        "openAICompatible": true,
        "openRouter": true,
        "openRouterModelId": "inception/mercury-2.5-preview",
        "openRouterPromotionalPricing": {
          "input": 0.04,
          "cachedInput": 0.004,
          "output": 0.15,
          "discount": "80%"
        }
      }
    },
    {
      "id": "ibm-granite42-8b-openrouter",
      "name": "IBM Granite 4.2 8B (OpenRouter)",
      "provider": "ibm",
      "tier": "Budget",
      "input": 0.1,
      "output": 0.15,
      "context": "128K",
      "deprecated": false,
      "note": "Official IBM open-weight dense reasoning model with tool calling and selectable thinking modes. Price is OpenRouter-hosted pricing, not IBM watsonx pricing; current first-party IBM-hosted API availability was not confirmed.",
      "availability": {
        "openWeights": true,
        "license": "Apache-2.0",
        "openRouter": true,
        "openRouterModelId": "ibm-granite/granite-4.2-8b",
        "ibmHostedApiConfirmed": false
      }
    },
    {
      "id": "zai-glm53-flash",
      "name": "GLM-5.3 Flash",
      "provider": "zai",
      "tier": "Budget",
      "input": 0.075,
      "output": 0.25,
      "context": "1M",
      "deprecated": false,
      "cachedInput": 0.015,
      "listInput": 0.15,
      "listCachedInput": 0.03,
      "listOutput": 0.5,
      "promotionStart": "2026-08-26",
      "promotionExpiry": "2026-09-09",
      "promotionEligibility": "Z.ai promotional rate through Sep 9, 2026 24:00 UTC+8; OpenRouter provider pricing may vary",
      "note": "OpenRouter ID z-ai/glm-5.3-flash. Native multimodal 320B/18B-active open-weight model under MIT; tool calling and structured outputs available on documented API surfaces. Ox Alpha alias is not confirmed in Z.ai primary release material."
    }
  ],
  "mediaModels": [
    {
      "id": "xai-grok-imagine-image-quality",
      "name": "Grok Imagine Image Quality",
      "provider": "xAI",
      "providerSlug": "xai",
      "modelId": "grok-imagine-image-quality",
      "modality": "image",
      "priceUnit": "per-image",
      "inputImage": 0.01,
      "outputByResolution": {
        "1k": 0.05,
        "2k": 0.07
      },
      "deprecated": true,
      "deprecatedDate": "2026-11-02",
      "replacement": "xai-grok-imagine-image-2",
      "redirectBehavior": {
        "modelId": "grok-imagine-image-2.0",
        "quality": "low",
        "endpointFailure": false
      },
      "verified": "Sep 7, 2026",
      "note": "The slug continues resolving after retirement; xAI serves requests with grok-imagine-image-2.0 at quality=low. Migrate explicitly to choose quality."
    },
    {
      "id": "xai-grok-imagine-image-2",
      "name": "Grok Imagine Image 2.0",
      "provider": "xAI",
      "providerSlug": "xai",
      "modelId": "grok-imagine-image-2.0",
      "modality": "image",
      "priceUnit": "per-image",
      "inputImage": 0.01,
      "outputByResolution": {
        "1k-low": 0.04,
        "2k-low": 0.06,
        "1k-medium": 0.06,
        "2k-medium": 0.08
      },
      "defaultQuality": "auto",
      "verified": "Sep 7, 2026",
      "note": "Auto currently serves low for generation and medium for editing; billing follows the quality served."
    }
  ]
}