{
  "updatedAt": "2026-07-10",
  "source": "ComparEdge verified model pricing",
  "url": "https://comparedge.com/llm-calculator",
  "license": "CC-BY-4.0",
  "pricingUnit": "USD per 1M tokens",
  "note": "input/output are USD per 1,000,000 tokens. inputCached is the discounted rate for cache-hit input where a provider offers it. contextK is the context window in thousands of tokens. Batch APIs (hasBatch) typically halve token cost.",
  "generated": "2026-07-11T05:31:04.044Z",
  "providerCount": 17,
  "modelCount": 69,
  "providers": [
    {
      "slug": "openai",
      "name": "OpenAI",
      "color": "#10a37f",
      "note": "Batch API: 50% discount. Cached input priced separately. Pro tiers (5.5-pro/5.4-pro) have no cached-input rate.",
      "hasBatch": true,
      "models": [
        {
          "id": "gpt-5.6-sol",
          "name": "GPT-5.6 Sol",
          "input": 5,
          "output": 30,
          "inputCached": 0.5,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "gpt-5.6-terra",
          "name": "GPT-5.6 Terra",
          "input": 2.5,
          "output": 15,
          "inputCached": 0.25,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "gpt-5.6-luna",
          "name": "GPT-5.6 Luna",
          "input": 1,
          "output": 6,
          "inputCached": 0.1,
          "contextK": 1000,
          "tier": "standard"
        },
        {
          "id": "gpt-5.5",
          "name": "GPT-5.5",
          "input": 5,
          "output": 30,
          "inputCached": 0.5,
          "contextK": 400,
          "tier": "flagship"
        },
        {
          "id": "gpt-5.4",
          "name": "GPT-5.4",
          "input": 2.5,
          "output": 15,
          "inputCached": 0.25,
          "contextK": 400,
          "tier": "flagship"
        },
        {
          "id": "gpt-5.4-mini",
          "name": "GPT-5.4 Mini",
          "input": 0.75,
          "output": 4.5,
          "inputCached": 0.075,
          "contextK": 400,
          "tier": "fast"
        },
        {
          "id": "gpt-5.4-nano",
          "name": "GPT-5.4 Nano",
          "input": 0.2,
          "output": 1.25,
          "inputCached": 0.02,
          "contextK": 400,
          "tier": "fast"
        },
        {
          "id": "gpt-5",
          "name": "GPT-5",
          "input": 1.25,
          "output": 10,
          "inputCached": 0.125,
          "contextK": 400,
          "tier": "standard"
        },
        {
          "id": "gpt-5-mini",
          "name": "GPT-5 Mini",
          "input": 0.25,
          "output": 2,
          "inputCached": 0.025,
          "contextK": 400,
          "tier": "fast"
        },
        {
          "id": "o3",
          "name": "o3",
          "input": 2,
          "output": 8,
          "inputCached": 0.5,
          "contextK": 200,
          "tier": "reasoning"
        },
        {
          "id": "o4-mini",
          "name": "o4-mini",
          "input": 1.1,
          "output": 4.4,
          "inputCached": 0.275,
          "contextK": 200,
          "tier": "reasoning"
        },
        {
          "id": "gpt-4.1",
          "name": "GPT-4.1",
          "input": 2,
          "output": 8,
          "inputCached": 0.5,
          "contextK": 1000,
          "tier": "standard"
        }
      ]
    },
    {
      "slug": "anthropic",
      "name": "Anthropic",
      "color": "#d97706",
      "note": "Batch API: 50% discount. Prompt caching available on all models. Sonnet 5 introductory pricing ($2/$10) runs through Aug 31, 2026.",
      "hasBatch": true,
      "models": [
        {
          "id": "claude-opus-4-8",
          "name": "Claude Opus 4.8",
          "input": 5,
          "output": 25,
          "inputCached": 0.5,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "claude-fable-5",
          "name": "Claude Fable 5",
          "input": 10,
          "output": 50,
          "inputCached": 1,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "claude-sonnet-5",
          "name": "Claude Sonnet 5",
          "input": 2,
          "output": 10,
          "inputCached": 0.2,
          "contextK": 1000,
          "tier": "standard"
        },
        {
          "id": "claude-haiku-4-5",
          "name": "Claude Haiku 4.5",
          "input": 1,
          "output": 5,
          "inputCached": 0.1,
          "contextK": 200,
          "tier": "fast"
        },
        {
          "id": "claude-opus-4-5",
          "name": "Claude Opus 4.5",
          "input": 5,
          "output": 25,
          "inputCached": 0.5,
          "contextK": 1000,
          "tier": "flagship"
        }
      ]
    },
    {
      "slug": "google",
      "name": "Google",
      "color": "#4285f4",
      "note": "Prices shown for prompts ≤200K tokens; prompts >200K cost more. Free tier available in Google AI Studio.",
      "hasBatch": true,
      "models": [
        {
          "id": "gemini-3.1-pro",
          "name": "Gemini 3.1 Pro",
          "input": 2,
          "output": 12,
          "inputCached": 0.2,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "gemini-3.5-flash",
          "name": "Gemini 3.5 Flash",
          "input": 1.5,
          "output": 9,
          "inputCached": 0.15,
          "contextK": 1000,
          "tier": "standard"
        },
        {
          "id": "gemini-3-flash",
          "name": "Gemini 3 Flash",
          "input": 0.5,
          "output": 3,
          "inputCached": 0.05,
          "contextK": 1000,
          "tier": "fast"
        },
        {
          "id": "gemini-3.1-flash-lite",
          "name": "Gemini 3.1 Flash-Lite",
          "input": 0.25,
          "output": 1.5,
          "inputCached": 0.025,
          "contextK": 1000,
          "tier": "fast"
        },
        {
          "id": "gemini-2.5-pro",
          "name": "Gemini 2.5 Pro",
          "input": 1.25,
          "output": 10,
          "inputCached": 0.125,
          "contextK": 1000,
          "tier": "standard"
        },
        {
          "id": "gemini-2.5-flash",
          "name": "Gemini 2.5 Flash",
          "input": 0.3,
          "output": 2.5,
          "inputCached": 0.03,
          "contextK": 1000,
          "tier": "fast"
        },
        {
          "id": "gemini-2.5-flash-lite",
          "name": "Gemini 2.5 Flash-Lite",
          "input": 0.1,
          "output": 0.4,
          "inputCached": 0.01,
          "contextK": 1000,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "deepseek",
      "name": "DeepSeek",
      "color": "#4d6bfe",
      "note": "Cache-hit input is far cheaper (V4-Flash $0.0028, V4-Pro $0.003625). Off-peak discounts apply. 1M context.",
      "hasBatch": false,
      "models": [
        {
          "id": "deepseek-v4-pro",
          "name": "DeepSeek-V4-Pro",
          "input": 0.435,
          "output": 0.87,
          "inputCached": 0.003625,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "deepseek-v4-flash",
          "name": "DeepSeek-V4-Flash",
          "input": 0.14,
          "output": 0.28,
          "inputCached": 0.0028,
          "contextK": 1000,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "amazon",
      "name": "Amazon",
      "color": "#ff9900",
      "note": "Amazon Nova on Bedrock. Batch and provisioned throughput available.",
      "hasBatch": true,
      "models": [
        {
          "id": "nova-premier",
          "name": "Amazon Nova Premier",
          "input": 2.5,
          "output": 12.5,
          "contextK": 300,
          "tier": "flagship"
        },
        {
          "id": "nova-pro",
          "name": "Amazon Nova Pro",
          "input": 0.8,
          "output": 3.2,
          "contextK": 300,
          "tier": "standard"
        },
        {
          "id": "nova-lite",
          "name": "Amazon Nova Lite",
          "input": 0.06,
          "output": 0.24,
          "contextK": 300,
          "tier": "fast"
        },
        {
          "id": "nova-micro",
          "name": "Amazon Nova Micro",
          "input": 0.035,
          "output": 0.14,
          "contextK": 128,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "groq",
      "name": "Groq",
      "color": "#f55036",
      "note": "Ultra-fast LPU inference for open-weight models. Pay-as-you-go, no subscription.",
      "hasBatch": false,
      "models": [
        {
          "id": "llama-4-maverick",
          "name": "Llama 4 Maverick",
          "input": 0.5,
          "output": 0.77,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "llama-4-scout",
          "name": "Llama 4 Scout",
          "input": 0.11,
          "output": 0.34,
          "contextK": 1000,
          "tier": "fast"
        },
        {
          "id": "gpt-oss-120b",
          "name": "GPT-OSS 120B",
          "input": 0.075,
          "output": 0.3,
          "contextK": 128,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "mistral",
      "name": "Mistral AI",
      "color": "#fa520f",
      "note": "Batch API: 50% discount. Prompt caching available. Consumer plans (Vibe) priced separately.",
      "hasBatch": true,
      "models": [
        {
          "id": "mistral-large-3",
          "name": "Mistral Large 3",
          "input": 0.5,
          "output": 1.5,
          "contextK": 256,
          "tier": "flagship"
        },
        {
          "id": "mistral-medium-3.5",
          "name": "Mistral Medium 3.5",
          "input": 1.5,
          "output": 7.5,
          "inputCached": 0.15,
          "contextK": 256,
          "tier": "flagship"
        },
        {
          "id": "magistral-medium",
          "name": "Magistral Medium",
          "input": 2,
          "output": 5,
          "contextK": 128,
          "tier": "reasoning"
        },
        {
          "id": "codestral",
          "name": "Codestral",
          "input": 0.3,
          "output": 0.9,
          "contextK": 256,
          "tier": "fast"
        },
        {
          "id": "mistral-small-4",
          "name": "Mistral Small 4",
          "input": 0.15,
          "output": 0.6,
          "contextK": 128,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "xai",
      "name": "xAI",
      "color": "#000000",
      "note": "Grok 4.5 is the reasoning flagship; Grok 4.3 is the non-reasoning flagship. Multi-agent variants also available. Prices per 1M tokens.",
      "hasBatch": false,
      "models": [
        {
          "id": "grok-4.5",
          "name": "Grok 4.5",
          "input": 2,
          "output": 6,
          "inputCached": 0.5,
          "contextK": 500,
          "tier": "reasoning"
        },
        {
          "id": "grok-4.3",
          "name": "Grok 4.3",
          "input": 1.25,
          "output": 2.5,
          "inputCached": 0.2,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "grok-build-0.1",
          "name": "Grok Build 0.1",
          "input": 1,
          "output": 2,
          "contextK": 256,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "minimax",
      "name": "MiniMax",
      "color": "#e2231a",
      "note": "MiniMax M3 is the current flagship (M2 retired). Prices per 1M tokens.",
      "hasBatch": false,
      "models": [
        {
          "id": "minimax-m3",
          "name": "MiniMax M3",
          "input": 0.3,
          "output": 1.2,
          "inputCached": 0.06,
          "contextK": 1000,
          "tier": "flagship"
        }
      ]
    },
    {
      "slug": "moonshot",
      "name": "Moonshot AI",
      "color": "#000000",
      "note": "Kimi K2 family. Cached input priced separately.",
      "hasBatch": false,
      "models": [
        {
          "id": "kimi-k2-6",
          "name": "Kimi K2.6",
          "input": 0.95,
          "output": 4,
          "inputCached": 0.16,
          "contextK": 256,
          "tier": "flagship"
        },
        {
          "id": "kimi-k2-0905",
          "name": "Kimi K2 0905",
          "input": 0.6,
          "output": 2.5,
          "inputCached": 0.15,
          "contextK": 256,
          "tier": "flagship"
        },
        {
          "id": "kimi-k2-thinking",
          "name": "Kimi K2 Thinking",
          "input": 0.6,
          "output": 2.5,
          "inputCached": 0.15,
          "contextK": 256,
          "tier": "reasoning"
        },
        {
          "id": "kimi-k2-turbo",
          "name": "Kimi K2 Turbo",
          "input": 1.15,
          "output": 8,
          "inputCached": 0.15,
          "contextK": 256,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "qwen",
      "name": "Qwen (Alibaba)",
      "color": "#615ced",
      "note": "Qwen3.7 Max is the proprietary flagship; Qwen3.5 397B A17B is open-weight (price = median across hosts). Prices rise above 256K context.",
      "hasBatch": false,
      "models": [
        {
          "id": "qwen3.7-max",
          "name": "Qwen3.7 Max",
          "input": 2.5,
          "output": 7.5,
          "inputCached": 0.25,
          "contextK": 1000,
          "tier": "flagship"
        },
        {
          "id": "qwen3.6-plus",
          "name": "Qwen3.6 Plus",
          "input": 0.5,
          "output": 3,
          "contextK": 256,
          "tier": "flagship"
        },
        {
          "id": "qwen3.5-397b",
          "name": "Qwen3.5 397B A17B",
          "input": 0.6,
          "output": 3.6,
          "contextK": 262,
          "tier": "standard"
        }
      ]
    },
    {
      "slug": "zai",
      "name": "Z.ai",
      "color": "#4f6bed",
      "note": "GLM family (Zhipu / Z.ai). Cache-hit pricing available. Prices per 1M tokens.",
      "hasBatch": false,
      "models": [
        {
          "id": "glm-5.2",
          "name": "GLM-5.2",
          "input": 1.4,
          "output": 4.4,
          "inputCached": 0.26,
          "contextK": 1000,
          "tier": "flagship"
        }
      ]
    },
    {
      "slug": "together",
      "name": "Together AI",
      "color": "#6D28D9",
      "note": "Open-source model hosting. Prices per 1M tokens.",
      "models": [
        {
          "id": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
          "name": "Llama 3.3 70B Turbo",
          "input": 0.88,
          "output": 0.88,
          "contextK": 128,
          "tier": "standard"
        },
        {
          "id": "meta-llama/Llama-3.1-405B-Instruct-Turbo",
          "name": "Llama 3.1 405B Turbo",
          "input": 3.5,
          "output": 3.5,
          "contextK": 128,
          "tier": "flagship"
        },
        {
          "id": "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo",
          "name": "Llama 3.1 8B Turbo",
          "input": 0.18,
          "output": 0.18,
          "contextK": 128,
          "tier": "fast"
        },
        {
          "id": "mistralai/Mixtral-8x7B-Instruct-v0.1",
          "name": "Mixtral 8x7B",
          "input": 0.6,
          "output": 0.6,
          "contextK": 32,
          "tier": "fast"
        },
        {
          "id": "Qwen/Qwen2.5-72B-Instruct-Turbo",
          "name": "Qwen 2.5 72B Turbo",
          "input": 1.2,
          "output": 1.2,
          "contextK": 32,
          "tier": "standard"
        },
        {
          "id": "deepseek-ai/DeepSeek-R1",
          "name": "DeepSeek R1 (Together)",
          "input": 7,
          "output": 7,
          "contextK": 64,
          "tier": "reasoning"
        }
      ]
    },
    {
      "slug": "cerebras",
      "name": "Cerebras",
      "color": "#EF4444",
      "note": "Wafer-scale AI chips. Ultra-low latency inference.",
      "models": [
        {
          "id": "llama-4-scout-17b",
          "name": "Llama 4 Scout 17B",
          "input": 0.4,
          "output": 0.4,
          "contextK": 128,
          "tier": "fast"
        },
        {
          "id": "cerebras-llama3.3-70b",
          "name": "Llama 3.3 70B",
          "input": 0.85,
          "output": 1.2,
          "contextK": 128,
          "tier": "standard"
        },
        {
          "id": "cerebras-llama3.1-8b",
          "name": "Llama 3.1 8B",
          "input": 0.1,
          "output": 0.1,
          "contextK": 128,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "fireworks",
      "name": "Fireworks AI",
      "color": "#F97316",
      "note": "Fast inference for open-source models.",
      "models": [
        {
          "id": "accounts/fireworks/models/llama-v3p3-70b-instruct",
          "name": "Llama 3.3 70B",
          "input": 0.9,
          "output": 0.9,
          "contextK": 128,
          "tier": "standard"
        },
        {
          "id": "accounts/fireworks/models/llama-v3p1-405b-instruct",
          "name": "Llama 3.1 405B",
          "input": 3,
          "output": 3,
          "contextK": 128,
          "tier": "flagship"
        },
        {
          "id": "accounts/fireworks/models/qwen2p5-72b-instruct",
          "name": "Qwen 2.5 72B",
          "input": 0.9,
          "output": 0.9,
          "contextK": 32,
          "tier": "standard"
        },
        {
          "id": "accounts/fireworks/models/deepseek-r1",
          "name": "DeepSeek R1 (Fireworks)",
          "input": 3,
          "output": 8,
          "contextK": 64,
          "tier": "reasoning"
        },
        {
          "id": "accounts/fireworks/models/firefunction-v2",
          "name": "FireFunction v2",
          "input": 0.9,
          "output": 0.9,
          "contextK": 32,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "perplexity",
      "name": "Perplexity",
      "color": "#20808d",
      "note": "Sonar models add a per-request search fee that varies by search-context size, on top of token costs.",
      "hasBatch": false,
      "models": [
        {
          "id": "sonar-pro",
          "name": "Sonar Pro",
          "input": 3,
          "output": 15,
          "contextK": 200,
          "tier": "flagship"
        },
        {
          "id": "sonar-reasoning-pro",
          "name": "Sonar Reasoning Pro",
          "input": 2,
          "output": 8,
          "contextK": 128,
          "tier": "reasoning"
        },
        {
          "id": "sonar",
          "name": "Sonar",
          "input": 1,
          "output": 1,
          "contextK": 128,
          "tier": "fast"
        }
      ]
    },
    {
      "slug": "cohere",
      "name": "Cohere",
      "color": "#39594c",
      "note": "Command A family for enterprise RAG and agents. Fine-tuning and dedicated deployment available.",
      "hasBatch": false,
      "models": [
        {
          "id": "command-a",
          "name": "Command A",
          "input": 2.5,
          "output": 10,
          "contextK": 256,
          "tier": "flagship"
        },
        {
          "id": "command-r7b",
          "name": "Command R7B",
          "input": 0.0375,
          "output": 0.15,
          "contextK": 128,
          "tier": "fast"
        }
      ]
    }
  ]
}