{
  "name": "Best AI model for every job",
  "url": "https://pulse.cafiyn.com/free/best-ai-models",
  "asOf": "2026-10-23",
  "license": "CC BY 4.0",
  "attribution": "Cafiyn Pulse, https://pulse.cafiyn.com/free/best-ai-models",
  "rules": {
    "text": "Text picks come from the routing engine: a cheap model for volume, a workhorse for routine work and a stronger model for hard cases, chosen by monthly call volume.",
    "media": "Lowest cost is the cheapest current model. Best value is the cheapest current model outside our basic tier. Premium is the cheapest current model in our premium tier. Tiers are our grouping by price band and the job each provider positions a model for, not a benchmark score.",
    "current": "Only current models are picked. Legacy, retired and delisted models are left out unless nothing else is selected."
  },
  "referenceWorkload": {
    "text": {
      "usersPerDay": 1000,
      "requestsPerUserPerDay": 5,
      "inputTokens": 1500,
      "outputTokens": 500
    },
    "tts": {
      "minutesPerDay": 500
    },
    "stt": {
      "minutesPerDay": 500
    },
    "image": {
      "imagesPerDay": 200
    },
    "video": {
      "secondsPerDay": 60
    },
    "daysPerMonth": 30
  },
  "rosterSize": 92,
  "categories": [
    {
      "category": "text",
      "name": "Text and chat",
      "page": "https://pulse.cafiyn.com/free/best-ai-models#text",
      "unit": "per million tokens, blended",
      "reference": "1,000 users a day making five requests each, 1,500 tokens in and 500 out",
      "picks": [
        {
          "roles": [
            "lowest"
          ],
          "id": "groq:gpt-oss-20b",
          "name": "GPT OSS 20B (Groq)",
          "provider": "Groq",
          "unitPrice": 0.13125,
          "monthlyUsdAtReference": 39.38,
          "notes": null
        },
        {
          "roles": [
            "value"
          ],
          "id": "anthropic:haiku-5.5",
          "name": "Claude Haiku 5.5",
          "provider": "Anthropic",
          "unitPrice": 0.2,
          "monthlyUsdAtReference": 60,
          "notes": "Prompts over 100,000 tokens cost $0.50 input and $2.50 output. Counts about 30 percent more tokens than Haiku 4.5 for the same text"
        },
        {
          "roles": [
            "premium"
          ],
          "id": "anthropic:sonnet-5.5",
          "name": "Claude Sonnet 5.5",
          "provider": "Anthropic",
          "unitPrice": 4,
          "monthlyUsdAtReference": 1200,
          "notes": "Cache reads $0.10 (cut from $0.20 on 7 October 2026). Uses Sonnet 5 tokenizer: about 30% more tokens than Sonnet 4.6, 4.5 or Haiku 4.5 for the same text, so compare on token counts, not rates"
        }
      ],
      "current": [
        {
          "id": "groq:gpt-oss-20b",
          "name": "GPT OSS 20B (Groq)",
          "provider": "Groq",
          "tier": "basic",
          "unitPrice": 0.13125,
          "monthlyUsdAtReference": 39.38,
          "notes": null
        },
        {
          "id": "anthropic:haiku-5.5",
          "name": "Claude Haiku 5.5",
          "provider": "Anthropic",
          "tier": "standard",
          "unitPrice": 0.2,
          "monthlyUsdAtReference": 60,
          "notes": "Prompts over 100,000 tokens cost $0.50 input and $2.50 output. Counts about 30 percent more tokens than Haiku 4.5 for the same text"
        },
        {
          "id": "openai:gpt-6-luna",
          "name": "GPT-6 Luna",
          "provider": "OpenAI",
          "tier": "standard",
          "unitPrice": 0.2,
          "monthlyUsdAtReference": 60,
          "notes": "Rate shown is for prompts up to 272K input tokens. Above that, $0.20 input and $0.75 output"
        },
        {
          "id": "deepseek:v4-flash",
          "name": "DeepSeek V4.1 Flash",
          "provider": "DeepSeek",
          "tier": "standard",
          "unitPrice": 0.2625,
          "monthlyUsdAtReference": 78.75,
          "notes": "Off peak rate. Peak (01:00 to 04:00 and 06:00 to 10:00 UTC, Mon to Fri) is 2x. API name deepseek-flash; the legacy name deepseek-v4-flash is still accepted and billed at this price"
        },
        {
          "id": "groq:gpt-oss-120b",
          "name": "GPT OSS 120B (Groq)",
          "provider": "Groq",
          "tier": "standard",
          "unitPrice": 0.2625,
          "monthlyUsdAtReference": 78.75,
          "notes": null
        },
        {
          "id": "openai:gpt-5.6-luna",
          "name": "GPT-5.6 Luna",
          "provider": "OpenAI",
          "tier": "standard",
          "unitPrice": 0.45,
          "monthlyUsdAtReference": 135,
          "notes": null
        },
        {
          "id": "mistral:large",
          "name": "Mistral Large",
          "provider": "Mistral",
          "tier": "premium",
          "unitPrice": 0.75,
          "monthlyUsdAtReference": 225,
          "notes": null
        },
        {
          "id": "google:gemini-3.5-flash-lite",
          "name": "Gemini 3.5 Flash Lite",
          "provider": "Google",
          "tier": "basic",
          "unitPrice": 0.85,
          "monthlyUsdAtReference": 255,
          "notes": null
        },
        {
          "id": "deepseek:v4-pro",
          "name": "DeepSeek V4 Pro",
          "provider": "DeepSeek",
          "tier": "premium",
          "unitPrice": 0.99,
          "monthlyUsdAtReference": 297,
          "notes": "Off peak rate. Peak (01:00 to 04:00 and 06:00 to 10:00 UTC, Mon to Fri) is 2x"
        },
        {
          "id": "perplexity:sonar",
          "name": "Sonar (web aware)",
          "provider": "Perplexity",
          "tier": "standard",
          "unitPrice": 1,
          "monthlyUsdAtReference": 300,
          "notes": "Plus a per request search fee of $5 to $14 per 1,000 requests by context size"
        },
        {
          "id": "meta:llama-3.3-70b",
          "name": "Llama 3.3 70B (Together)",
          "provider": "Together",
          "tier": "standard",
          "unitPrice": 1.04,
          "monthlyUsdAtReference": 312,
          "notes": "Groq no longer publishes a per token price for this model (contact sales), so this is the Together rate"
        },
        {
          "id": "google:gemini-3.7-flash",
          "name": "Gemini 3.7 Flash",
          "provider": "Google",
          "tier": "standard",
          "unitPrice": 1.5,
          "monthlyUsdAtReference": 450,
          "notes": "Scheduled increase to $1.50 input and $7.50 output on 1 Jan 2027"
        },
        {
          "id": "google:gemini-3.8-flash",
          "name": "Gemini 3.8 Flash",
          "provider": "Google",
          "tier": "standard",
          "unitPrice": 1.5,
          "monthlyUsdAtReference": 450,
          "notes": "Scheduled increase to $1.50 input and $7.50 output on 1 Jan 2027"
        },
        {
          "id": "xai:grok-4.3",
          "name": "Grok 4.3",
          "provider": "xAI",
          "tier": "standard",
          "unitPrice": 1.5625,
          "monthlyUsdAtReference": 468.75,
          "notes": "Rate doubles on prompts of 200k tokens or more"
        },
        {
          "id": "openai:gpt-5.4-mini",
          "name": "GPT-5.4 mini",
          "provider": "OpenAI",
          "tier": "standard",
          "unitPrice": 1.6875,
          "monthlyUsdAtReference": 506.25,
          "notes": null
        },
        {
          "id": "anthropic:haiku-4.5",
          "name": "Claude Haiku 4.5",
          "provider": "Anthropic",
          "tier": "standard",
          "unitPrice": 2,
          "monthlyUsdAtReference": 600,
          "notes": null
        },
        {
          "id": "xai:grok-4.6",
          "name": "Grok 4.6",
          "provider": "xAI",
          "tier": "premium",
          "unitPrice": 3,
          "monthlyUsdAtReference": 900,
          "notes": "Rate doubles on prompts of 200k tokens or more"
        },
        {
          "id": "xai:grok-4.7",
          "name": "Grok 4.7",
          "provider": "xAI",
          "tier": "premium",
          "unitPrice": 3,
          "monthlyUsdAtReference": 900,
          "notes": "Rate doubles on prompts of 200k tokens or more"
        },
        {
          "id": "google:gemini-3.5-flash",
          "name": "Gemini 3.5 Flash",
          "provider": "Google",
          "tier": "standard",
          "unitPrice": 3.375,
          "monthlyUsdAtReference": 1012.5,
          "notes": null
        },
        {
          "id": "perplexity:sonar-reasoning-pro",
          "name": "Sonar Reasoning Pro",
          "provider": "Perplexity",
          "tier": "premium",
          "unitPrice": 3.5,
          "monthlyUsdAtReference": 1050,
          "notes": "Plus a per request search fee of $5 to $14 per 1,000 requests by context size"
        },
        {
          "id": "anthropic:sonnet-5",
          "name": "Claude Sonnet 5",
          "provider": "Anthropic",
          "tier": "premium",
          "unitPrice": 4,
          "monthlyUsdAtReference": 1200,
          "notes": null
        },
        {
          "id": "anthropic:sonnet-5.5",
          "name": "Claude Sonnet 5.5",
          "provider": "Anthropic",
          "tier": "premium",
          "unitPrice": 4,
          "monthlyUsdAtReference": 1200,
          "notes": "Cache reads $0.10 (cut from $0.20 on 7 October 2026). Uses Sonnet 5 tokenizer: about 30% more tokens than Sonnet 4.6, 4.5 or Haiku 4.5 for the same text, so compare on token counts, not rates"
        },
        {
          "id": "openai:gpt-6-sol",
          "name": "GPT-6 Sol",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 4,
          "monthlyUsdAtReference": 1200,
          "notes": "Rate shown is for prompts up to 272K input tokens. Above that, $4 input and $15 output"
        },
        {
          "id": "openai:gpt-6.1-sol",
          "name": "GPT-6.1 Sol",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 4,
          "monthlyUsdAtReference": 1200,
          "notes": "Rate shown is for prompts up to 272K input tokens. Above that, $4 input and $15 output. Cached input $0.10, half the rate on GPT-6 Sol"
        },
        {
          "id": "google:gemini-3.1-pro",
          "name": "Gemini 3.1 Pro",
          "provider": "Google",
          "tier": "premium",
          "unitPrice": 4.5,
          "monthlyUsdAtReference": 1350,
          "notes": null
        },
        {
          "id": "openai:gpt-5.6-terra",
          "name": "GPT-5.6 Terra",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 4.5,
          "monthlyUsdAtReference": 1350,
          "notes": null
        },
        {
          "id": "openai:gpt-5.4",
          "name": "GPT-5.4",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 5.625,
          "monthlyUsdAtReference": 1687.5,
          "notes": null
        },
        {
          "id": "perplexity:sonar-pro",
          "name": "Sonar Pro (web aware)",
          "provider": "Perplexity",
          "tier": "premium",
          "unitPrice": 6,
          "monthlyUsdAtReference": 1800,
          "notes": "Plus a per request search fee of $5 to $14 per 1,000 requests by context size"
        },
        {
          "id": "anthropic:opus-5.5",
          "name": "Claude Opus 5.5",
          "provider": "Anthropic",
          "tier": "premium",
          "unitPrice": 8,
          "monthlyUsdAtReference": 2400,
          "notes": "Cache hits are 0.05x base input ($0.20), not the usual 0.1x. Thinking cannot be disabled and tool_choice any/tool return 400"
        },
        {
          "id": "openai:gpt-5.6-sol",
          "name": "GPT-5.6 Sol",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 8,
          "monthlyUsdAtReference": 2400,
          "notes": "OpenAI describes this as promotional pricing, available at least through 21 Nov 2026"
        },
        {
          "id": "anthropic:opus-5",
          "name": "Claude Opus 5",
          "provider": "Anthropic",
          "tier": "premium",
          "unitPrice": 10,
          "monthlyUsdAtReference": 3000,
          "notes": null
        },
        {
          "id": "openai:gpt-5.5",
          "name": "GPT-5.5",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 11.25,
          "monthlyUsdAtReference": 3375,
          "notes": null
        },
        {
          "id": "anthropic:fable-5.1",
          "name": "Claude Fable 5.1",
          "provider": "Anthropic",
          "tier": "premium",
          "unitPrice": 20,
          "monthlyUsdAtReference": 6000,
          "notes": null
        },
        {
          "id": "openai:gpt-6-astra",
          "name": "GPT-6 Astra",
          "provider": "OpenAI",
          "tier": "premium",
          "unitPrice": 20,
          "monthlyUsdAtReference": 6000,
          "notes": "Rate shown is for prompts up to 272K input tokens. Above that, $20 input and $75 output"
        }
      ]
    },
    {
      "category": "tts",
      "name": "Text to speech",
      "page": "https://pulse.cafiyn.com/free/best-ai-models#text-to-speech",
      "unit": "a minute",
      "reference": "500 minutes of generated speech a day",
      "picks": [
        {
          "roles": [
            "lowest",
            "value"
          ],
          "id": "azure:neural-tts",
          "name": "Azure Neural TTS",
          "provider": "Azure",
          "unitPrice": 0.015,
          "monthlyUsdAtReference": 225,
          "notes": "$15 per million characters, East US pay as you go"
        },
        {
          "roles": [
            "premium"
          ],
          "id": "elevenlabs:multilingual-v2",
          "name": "ElevenLabs Multilingual v2",
          "provider": "ElevenLabs",
          "unitPrice": 0.08,
          "monthlyUsdAtReference": 1200,
          "notes": "$0.08 per 1,000 characters, about one minute of speech"
        }
      ],
      "current": [
        {
          "id": "azure:neural-tts",
          "name": "Azure Neural TTS",
          "provider": "Azure",
          "tier": "standard",
          "unitPrice": 0.015,
          "monthlyUsdAtReference": 225,
          "notes": "$15 per million characters, East US pay as you go"
        },
        {
          "id": "deepgram:aura",
          "name": "Deepgram Aura",
          "provider": "Deepgram",
          "tier": "standard",
          "unitPrice": 0.015,
          "monthlyUsdAtReference": 225,
          "notes": null
        },
        {
          "id": "google:tts",
          "name": "Google Cloud TTS (Neural2)",
          "provider": "Google",
          "tier": "standard",
          "unitPrice": 0.016,
          "monthlyUsdAtReference": 240,
          "notes": null
        },
        {
          "id": "elevenlabs:turbo-v2",
          "name": "ElevenLabs Flash / Turbo",
          "provider": "ElevenLabs",
          "tier": "standard",
          "unitPrice": 0.04,
          "monthlyUsdAtReference": 600,
          "notes": "$0.04 per 1,000 characters, about one minute of speech"
        },
        {
          "id": "elevenlabs:multilingual-v2",
          "name": "ElevenLabs Multilingual v2",
          "provider": "ElevenLabs",
          "tier": "premium",
          "unitPrice": 0.08,
          "monthlyUsdAtReference": 1200,
          "notes": "$0.08 per 1,000 characters, about one minute of speech"
        },
        {
          "id": "google:tts-studio",
          "name": "Google Cloud TTS (Studio)",
          "provider": "Google",
          "tier": "premium",
          "unitPrice": 0.16,
          "monthlyUsdAtReference": 2400,
          "notes": null
        }
      ]
    },
    {
      "category": "stt",
      "name": "Speech to text",
      "page": "https://pulse.cafiyn.com/free/best-ai-models#speech-to-text",
      "unit": "a minute",
      "reference": "500 minutes of audio transcribed a day",
      "picks": [
        {
          "roles": [
            "lowest",
            "value",
            "premium"
          ],
          "id": "assemblyai:universal-2",
          "name": "AssemblyAI Universal-2",
          "provider": "AssemblyAI",
          "unitPrice": 0.0025,
          "monthlyUsdAtReference": 37.5,
          "notes": "$0.15 an hour. Universal-3.5 Pro is $0.21 an hour"
        }
      ],
      "current": [
        {
          "id": "assemblyai:universal-2",
          "name": "AssemblyAI Universal-2",
          "provider": "AssemblyAI",
          "tier": "premium",
          "unitPrice": 0.0025,
          "monthlyUsdAtReference": 37.5,
          "notes": "$0.15 an hour. Universal-3.5 Pro is $0.21 an hour"
        },
        {
          "id": "deepgram:nova-3",
          "name": "Deepgram Nova 3",
          "provider": "Deepgram",
          "tier": "premium",
          "unitPrice": 0.0043,
          "monthlyUsdAtReference": 64.5,
          "notes": "Prerecorded audio, monolingual. Streaming lists $0.0048 a minute, discounted from $0.0077"
        },
        {
          "id": "google:stt",
          "name": "Google Cloud STT",
          "provider": "Google",
          "tier": "standard",
          "unitPrice": 0.016,
          "monthlyUsdAtReference": 240,
          "notes": null
        },
        {
          "id": "azure:stt",
          "name": "Azure Speech to Text",
          "provider": "Azure",
          "tier": "standard",
          "unitPrice": 0.017,
          "monthlyUsdAtReference": 255,
          "notes": null
        }
      ]
    },
    {
      "category": "image",
      "name": "Image generation",
      "page": "https://pulse.cafiyn.com/free/best-ai-models#image",
      "unit": "an image",
      "reference": "200 images a day",
      "picks": [
        {
          "roles": [
            "lowest",
            "value"
          ],
          "id": "openai:gpt-image-2.5",
          "name": "GPT Image 2.5 (medium)",
          "provider": "OpenAI",
          "unitPrice": 0.0132,
          "monthlyUsdAtReference": 79.2,
          "notes": "Priced per token: about 439 output tokens for a 1024x1024 medium image at $30 per million, plus prompt tokens. High quality is about $0.053"
        },
        {
          "roles": [
            "premium"
          ],
          "id": "black-forest:flux-2-pro",
          "name": "FLUX.2 [pro]",
          "provider": "BFL",
          "unitPrice": 0.03,
          "monthlyUsdAtReference": 180,
          "notes": "Per 1 megapixel image"
        }
      ],
      "current": [
        {
          "id": "openai:gpt-image-2.5",
          "name": "GPT Image 2.5 (medium)",
          "provider": "OpenAI",
          "tier": "standard",
          "unitPrice": 0.0132,
          "monthlyUsdAtReference": 79.2,
          "notes": "Priced per token: about 439 output tokens for a 1024x1024 medium image at $30 per million, plus prompt tokens. High quality is about $0.053"
        },
        {
          "id": "black-forest:flux-2-klein-4b",
          "name": "FLUX.2 [klein] 4B",
          "provider": "BFL",
          "tier": "basic",
          "unitPrice": 0.014,
          "monthlyUsdAtReference": 84,
          "notes": "Per 1 megapixel image"
        },
        {
          "id": "black-forest:flux-2-pro",
          "name": "FLUX.2 [pro]",
          "provider": "BFL",
          "tier": "premium",
          "unitPrice": 0.03,
          "monthlyUsdAtReference": 180,
          "notes": "Per 1 megapixel image"
        },
        {
          "id": "recraft:v3",
          "name": "Recraft v3",
          "provider": "Recraft",
          "tier": "premium",
          "unitPrice": 0.04,
          "monthlyUsdAtReference": 240,
          "notes": null
        },
        {
          "id": "stability:sd-3.5",
          "name": "Stable Diffusion 3.5",
          "provider": "Stability",
          "tier": "premium",
          "unitPrice": 0.065,
          "monthlyUsdAtReference": 390,
          "notes": null
        },
        {
          "id": "ideogram:v2",
          "name": "Ideogram v2",
          "provider": "Ideogram",
          "tier": "premium",
          "unitPrice": 0.08,
          "monthlyUsdAtReference": 480,
          "notes": "Not re-verified on 23 Oct 2026: Ideogram renders API prices client side. Price as of 2 Oct 2026"
        },
        {
          "id": "ideogram:v3",
          "name": "Ideogram v3",
          "provider": "Ideogram",
          "tier": "premium",
          "unitPrice": 0.09,
          "monthlyUsdAtReference": 540,
          "notes": "Not re-verified on 23 Oct 2026: Ideogram renders API prices client side. Price as of 2 Oct 2026"
        }
      ]
    },
    {
      "category": "video",
      "name": "Video generation",
      "page": "https://pulse.cafiyn.com/free/best-ai-models#video",
      "unit": "a second",
      "reference": "60 seconds of video a day",
      "picks": [
        {
          "roles": [
            "lowest",
            "value"
          ],
          "id": "runway:gen-4-turbo",
          "name": "Runway Gen-4 Turbo",
          "provider": "Runway",
          "unitPrice": 0.05,
          "monthlyUsdAtReference": 90,
          "notes": "5 credits a second at $0.01 a credit"
        },
        {
          "roles": [
            "premium"
          ],
          "id": "kling:3.0",
          "name": "Kling 3.0 (720p)",
          "provider": "Kling",
          "unitPrice": 0.084,
          "monthlyUsdAtReference": 151.2,
          "notes": "0.6 units a second at $0.14 a unit, without native audio"
        }
      ],
      "current": [
        {
          "id": "runway:gen-4-turbo",
          "name": "Runway Gen-4 Turbo",
          "provider": "Runway",
          "tier": "standard",
          "unitPrice": 0.05,
          "monthlyUsdAtReference": 90,
          "notes": "5 credits a second at $0.01 a credit"
        },
        {
          "id": "luma:dream-machine",
          "name": "Luma Ray3.2 (720p)",
          "provider": "Luma",
          "tier": "standard",
          "unitPrice": 0.06,
          "monthlyUsdAtReference": 108,
          "notes": "$0.30 per 5 second clip at 720p. A 10 second clip is $0.90"
        },
        {
          "id": "kling:3.0",
          "name": "Kling 3.0 (720p)",
          "provider": "Kling",
          "tier": "premium",
          "unitPrice": 0.084,
          "monthlyUsdAtReference": 151.2,
          "notes": "0.6 units a second at $0.14 a unit, without native audio"
        },
        {
          "id": "runway:gen-4.5",
          "name": "Runway Gen-4.5",
          "provider": "Runway",
          "tier": "premium",
          "unitPrice": 0.12,
          "monthlyUsdAtReference": 216,
          "notes": "12 credits a second at $0.01 a credit"
        }
      ]
    }
  ]
}