{
  "schema_version": 1,
  "generated_note": "All values were read from each provider's official page on the date in last_verified. A null value means we could not verify it from an official source; the field name is then listed in `unverified`. Never fill a null with a guess.",
  "providers": [
    {
      "slug": "groq",
      "name": "Groq",
      "category": "LLM inference",
      "website": "https://console.groq.com",
      "status": "free-tier",
      "free_type": "Recurring free plan",
      "summary": "Permanent Free plan with per-model daily caps on open-weight models (gpt-oss, Qwen, Whisper).",
      "free_models": [
        "openai/gpt-oss-120b",
        "openai/gpt-oss-20b",
        "qwen/qwen3.8-27b",
        "whisper-large-v3",
        "whisper-large-v3-turbo",
        "canopylabs/orpheus-v1-english",
        "meta-llama/llama-prompt-guard-2-86m"
      ],
      "free_quota": "Per-model daily caps, e.g. gpt-oss-120b: 1K requests/day and 200K tokens/day.",
      "rate_limits": {
        "example_model": "openai/gpt-oss-120b",
        "rpm": 30,
        "rpd": 1000,
        "tpm": 8000,
        "tpd": 200000,
        "other": "Whisper: 20 RPM, 2K RPD, 7.2K audio-sec/hour, 28.8K audio-sec/day"
      },
      "limits_table": [
        {
          "model": "openai/gpt-oss-120b",
          "rpm": "30",
          "rpd": "1K",
          "tpm": "8K",
          "tpd": "200K"
        },
        {
          "model": "openai/gpt-oss-20b",
          "rpm": "30",
          "rpd": "1K",
          "tpm": "8K",
          "tpd": "200K"
        },
        {
          "model": "qwen/qwen3.8-27b",
          "rpm": "30",
          "rpd": "1K",
          "tpm": "8K",
          "tpd": "200K"
        },
        {
          "model": "openai/gpt-oss-safeguard-20b",
          "rpm": "3",
          "rpd": "1K",
          "tpm": "2K",
          "tpd": "200K"
        },
        {
          "model": "whisper-large-v3 / -turbo",
          "rpm": "20",
          "rpd": "2K",
          "tpm": "—",
          "tpd": "— (7.2K audio-sec/h, 28.8K/day)"
        },
        {
          "model": "canopylabs/orpheus-v1-english (TTS)",
          "rpm": "10",
          "rpd": "100",
          "tpm": "1.2K",
          "tpd": "3.6K"
        }
      ],
      "card_required": "no",
      "card_note": "Billing FAQ: a payment method is needed only to upgrade from the Free tier to the Developer tier.",
      "trial_credits": null,
      "trial_note": "Billing FAQ: promotional credits are offered occasionally (hackathons, events). No standing sign-up credit.",
      "use_cases": [
        "chat",
        "coding-agent",
        "speech",
        "no-card"
      ],
      "sources": [
        {
          "label": "Rate limits (Free Plan tab)",
          "url": "https://console.groq.com/docs/rate-limits"
        },
        {
          "label": "Billing FAQs",
          "url": "https://console.groq.com/docs/billing-faqs"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "trial_credits"
      ],
      "notes": [
        "Limits apply per organization, not per user or key.",
        "Cached tokens do not count toward rate limits.",
        "Docs include setup guides for coding agents (OpenCode, Cline, Roo Code, Kilo Code, Factory Droid)."
      ],
      "best_free_model": "openai/gpt-oss-120b",
      "headline": "1,000 req/day"
    },
    {
      "slug": "google-gemini",
      "name": "Google Gemini API",
      "category": "LLM inference",
      "website": "https://ai.google.dev",
      "status": "free-tier",
      "free_type": "Recurring free tier",
      "summary": "Free tier on Flash, Flash-Lite, 2.5 Pro, Gemma 4 and embeddings; newest flagship (Gemini 4 Argon) is paid-only.",
      "free_models": [
        "gemini-3.8-flash",
        "gemini-3.6-flash",
        "gemini-3.5-flash-lite",
        "gemini-3.1-flash-lite",
        "gemini-3-flash-preview",
        "gemini-2.5-pro",
        "gemini-2.5-flash",
        "gemini-2.5-flash-lite",
        "gemini-embedding-2",
        "Gemma 4",
        "gemini-3.8-flash-tts"
      ],
      "free_quota": "Input & output tokens 'Free of charge' on eligible models; per-model RPM/TPM/RPD limits are shown only inside Google AI Studio. Grounding with Google Search: 500 RPD free (shared for Flash and Flash-Lite).",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Free-tier numbers are no longer published on the docs page — view them in AI Studio. RPD resets at midnight Pacific."
      },
      "card_required": "no",
      "card_note": "Rate-limits doc: Free tier qualification is 'Active project or free trial'; linking a billing account moves you to Tier 1.",
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "chat",
        "coding-agent",
        "embeddings",
        "speech",
        "no-card"
      ],
      "sources": [
        {
          "label": "Gemini API pricing",
          "url": "https://ai.google.dev/gemini-api/docs/pricing"
        },
        {
          "label": "Rate limits",
          "url": "https://ai.google.dev/gemini-api/docs/rate-limits"
        },
        {
          "label": "Deprecations",
          "url": "https://ai.google.dev/gemini-api/docs/deprecations"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits.rpm",
        "rate_limits.rpd",
        "rate_limits.tpm"
      ],
      "notes": [
        "Not free: Gemini 4 Argon, Gemini 3.1 Pro Preview, Nano Banana image models, Veo, Lyria.",
        "Free-tier content may be used to improve Google products (pricing page).",
        "Deprecations page: access to 2.5 models is being limited to users who used them before; new projects are pointed to 3.5 Flash-Lite / 3.8 Flash.",
        "The 9 Oct 2026 'free Gemini is Flash-Lite only' news concerns the Gemini consumer app, not the API."
      ],
      "best_free_model": "gemini-3.8-flash",
      "headline": "Free tokens; limits in AI Studio"
    },
    {
      "slug": "openrouter",
      "name": "OpenRouter (free models)",
      "category": "Gateway / router",
      "website": "https://openrouter.ai",
      "status": "free-tier",
      "free_type": "Recurring free models (:free variants)",
      "summary": "17 ':free' model variants today; 50 free requests/day per account, or 1,000/day once you have bought ≥10 credits.",
      "free_models": [
        "nvidia/nemotron-3-super-120b-a12b:free",
        "nvidia/nemotron-3-ultra-550b-a55b:free",
        "google/gemma-4-31b-it:free",
        "google/gemma-4-26b-a4b-it:free",
        "cohere/north-mini-code:free",
        "poolside/laguna-s-2.1:free",
        "poolside/laguna-xs-2.1:free",
        "minimax/minimax-m3.1-flash:free",
        "thinkingmachines/inkling:free",
        "thinkingmachines/inkling-small:free",
        "nvidia/nemotron-3.5-super-vl:free",
        "dots-studio/dots-3-note-preview:free",
        "+5 more"
      ],
      "free_quota": "Free-model requests: 50/day if you purchased <10 credits all-time; 1,000/day with ≥10 credits purchased.",
      "rate_limits": {
        "example_model": "any :free model",
        "rpm": 20,
        "rpd": 50,
        "tpm": null,
        "tpd": null,
        "other": "RPD rises to 1,000 after ≥10 credits purchased (all-time). Upstream providers may also 429.",
        "rpd_scope": "account",
        "scope_note": "Requests/day is one account-wide counter across all :free models (reported as free_model_daily_requests by GET /api/v1/key) — not a per-model limit. 20 requests/minute applies to :free model requests."
      },
      "limits_table": [
        {
          "model": "Credits purchased < 10",
          "rpm": "20",
          "rpd": "50",
          "tpm": "—",
          "tpd": "—"
        },
        {
          "model": "Credits purchased ≥ 10",
          "rpm": "20",
          "rpd": "1,000",
          "tpm": "—",
          "tpd": "—"
        }
      ],
      "card_required": "no",
      "card_note": "Free models work at zero purchased credits (lower 50 RPD tier). A negative balance can block free models too.",
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "chat",
        "coding-agent",
        "no-card"
      ],
      "sources": [
        {
          "label": "API limits",
          "url": "https://openrouter.ai/docs/api-reference/limits"
        },
        {
          "label": "Models API (filter ':free')",
          "url": "https://openrouter.ai/api/v1/models"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits.tpm"
      ],
      "notes": [
        "The daily free limit is account-wide: all :free models share one counter. It is not per model.",
        "Free model list counted from the public /api/v1/models endpoint on 2026-10-21 (17 of 468 models).",
        "Check remaining free requests via GET /api/v1/key → free_model_daily_requests."
      ],
      "best_free_model": "nemotron-3-super-120b:free",
      "headline": "50 req/day per account · 1,000 with ≥10 credits"
    },
    {
      "slug": "mistral",
      "name": "Mistral AI (La Plateforme)",
      "category": "LLM inference",
      "website": "https://console.mistral.ai",
      "status": "free-tier",
      "free_type": "Recurring free mode",
      "summary": "'Free mode' includes a monthly usage allowance; exact limits are only shown in your Admin Panel.",
      "free_models": [],
      "free_quota": null,
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Limits are requests/second + tokens/minute (+ monthly), per organization, listed in Admin Panel › API › Limits."
      },
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": "Le Chat/Vibe paid plans include monthly API credits ($15–$25.5/mo per pricing page) — not a free offer.",
      "use_cases": [
        "chat"
      ],
      "sources": [
        {
          "label": "Usage and limits",
          "url": "https://docs.mistral.ai/admin/billing-usage/usage-limits"
        },
        {
          "label": "Pricing",
          "url": "https://mistral.ai/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "free_quota",
        "rate_limits",
        "card_required",
        "free_models"
      ],
      "notes": [
        "Docs: 'Free mode lets you create API keys and use included monthly usage within the limits shown on the Limits page.'",
        "Numeric free limits are not published publicly; we will not guess them."
      ],
      "best_free_model": null,
      "headline": "Monthly allowance (unpublished)"
    },
    {
      "slug": "cohere",
      "name": "Cohere",
      "category": "LLM + embeddings + rerank",
      "website": "https://dashboard.cohere.com",
      "status": "free-tier",
      "free_type": "Free trial (evaluation) keys",
      "summary": "Free trial keys: 1,000 API calls per month, 20 req/min on Chat models, plus Embed and Rerank.",
      "free_models": [
        "Command A",
        "Command A+",
        "Command A Reasoning",
        "Command A Vision",
        "Command R+",
        "Command R",
        "Command R7B",
        "North Mini Code",
        "Embed",
        "Rerank"
      ],
      "free_quota": "1,000 API calls per month on trial keys.",
      "rate_limits": {
        "example_model": "Chat (any Command model)",
        "rpm": 20,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Embed: 2,000 inputs/min · Rerank: 10 req/min · Audio transcription: 5 req/min"
      },
      "limits_table": [
        {
          "model": "Chat (Command A / R / R7B / North Mini Code …)",
          "rpm": "20",
          "rpd": "—",
          "tpm": "—",
          "tpd": "1,000 calls / month"
        },
        {
          "model": "Embed",
          "rpm": "2,000 inputs/min",
          "rpd": "—",
          "tpm": "—",
          "tpd": "—"
        },
        {
          "model": "Rerank",
          "rpm": "10",
          "rpd": "—",
          "tpm": "—",
          "tpd": "—"
        },
        {
          "model": "Audio Transcriptions",
          "rpm": "5",
          "rpd": "—",
          "tpm": "—",
          "tpd": "—"
        }
      ],
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "chat",
        "embeddings",
        "rerank",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "Rate limits",
          "url": "https://docs.cohere.com/docs/rate-limits"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "card_required"
      ],
      "notes": [
        "Trial keys are 'evaluation keys (free but limited in usage)'.",
        "Newer model variants (e.g. Command A Reasoning) are limited like trial keys even on production keys."
      ],
      "best_free_model": "Command A",
      "headline": "1,000 calls/month"
    },
    {
      "slug": "cloudflare-workers-ai",
      "name": "Cloudflare Workers AI",
      "category": "LLM + multimodal inference",
      "website": "https://developers.cloudflare.com/workers-ai/",
      "status": "free-tier",
      "free_type": "Recurring daily allocation",
      "summary": "10,000 Neurons per day free on both Workers Free and Paid; some frontier models now require a paid billing method.",
      "free_models": [
        "@cf/meta/llama-3.3-70b-instruct-fp8-fast",
        "@cf/meta/llama-3.1-8b-instruct",
        "@cf/mistralai/mistral-small-3.1-24b-instruct",
        "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b",
        "+ most catalog models"
      ],
      "free_quota": "10,000 Neurons/day (resets 00:00 UTC). Above that: $0.011 per 1,000 Neurons on Workers Paid.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Daily Neuron budget is the main free limit."
      },
      "card_required": "unverified",
      "card_note": "Paid billing method is required for: kimi-k2.6, kimi-k2.7-code, glm-5.2, glm-5.3, glm-5.3-flash, deepseek-v4-flash-0731, deepseek-v4.1-flash-1017, deepseek-v4-pro-0813.",
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "chat",
        "embeddings",
        "image",
        "speech"
      ],
      "sources": [
        {
          "label": "Workers AI pricing",
          "url": "https://developers.cloudflare.com/workers-ai/platform/pricing/"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits",
        "card_required"
      ],
      "notes": [
        "Official page shows 'Last updated Oct 20, 2026'.",
        "Example rate: llama-3.3-70b-instruct-fp8-fast = 26,668 neurons per M input tokens, 204,805 per M output tokens."
      ],
      "best_free_model": "llama-3.3-70b-instruct",
      "headline": "10,000 Neurons/day"
    },
    {
      "slug": "github-models",
      "name": "GitHub Models",
      "category": "LLM inference",
      "website": "https://docs.github.com/en/github-models",
      "status": "retired",
      "free_type": "Retired",
      "summary": "Fully retired on July 30, 2026 — playground, catalog, inference API and BYOK are gone.",
      "free_models": [],
      "free_quota": null,
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Service no longer available."
      },
      "card_required": "n/a",
      "card_note": null,
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [],
      "sources": [
        {
          "label": "GitHub Models docs (retirement notice)",
          "url": "https://docs.github.com/en/github-models/use-github-models/prototyping-with-ai-models"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [],
      "notes": [
        "GitHub points users to Azure AI Foundry or GitHub Copilot instead.",
        "Was a separate service from GitHub Copilot."
      ],
      "best_free_model": null,
      "headline": "Retired Jul 30, 2026"
    },
    {
      "slug": "hugging-face",
      "name": "Hugging Face Inference Providers",
      "category": "Gateway / router",
      "website": "https://huggingface.co/docs/inference-providers",
      "status": "no-free-tier",
      "free_type": "No free credits for free accounts",
      "summary": "Free users get no monthly inference credits; PRO users and Team/Enterprise seats get $2.00/month.",
      "free_models": [],
      "free_quota": "Free users: none (credit purchase required). PRO: $2.00/month. Team/Enterprise: $2.00 per seat/month.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": null
      },
      "card_required": "yes",
      "card_note": "Free users must purchase credits (or upgrade to PRO) to use Inference Providers.",
      "trial_credits": null,
      "trial_note": "Hub hosting remains free (e.g. CPU Basic Spaces, ZeroGPU listed as FREE on the pricing page).",
      "use_cases": [],
      "sources": [
        {
          "label": "Inference Providers pricing & billing",
          "url": "https://huggingface.co/docs/inference-providers/pricing"
        },
        {
          "label": "Hugging Face pricing",
          "url": "https://huggingface.co/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits"
      ],
      "notes": [
        "Pass-through provider pricing with no Hugging Face markup.",
        "Credits do not apply when you use a custom provider key."
      ],
      "best_free_model": null,
      "headline": "No free credits"
    },
    {
      "slug": "cerebras",
      "name": "Cerebras Inference",
      "category": "LLM inference",
      "website": "https://cloud.cerebras.ai",
      "status": "trial-only",
      "free_type": "One-time trial credits",
      "summary": "No permanent free tier: $5 trial credits after adding a verified payment method, expiring after 30 days.",
      "free_models": [
        "gpt-oss-120b",
        "qwen-3.8-27b"
      ],
      "free_quota": "$5 credits, expire 30 days after grant.",
      "rate_limits": {
        "example_model": "gpt-oss-120b (Free Trial)",
        "rpm": 5,
        "rpd": null,
        "tpm": 30000,
        "tpd": 1000000,
        "other": "Total TPM (incl. cached) 90K · TPH 1M"
      },
      "limits_table": [
        {
          "model": "gpt-oss-120b (Free Trial)",
          "rpm": "5",
          "rpd": "—",
          "tpm": "30K uncached / 90K total",
          "tpd": "1M (TPH 1M)"
        },
        {
          "model": "qwen-3.8-27b (Free Trial)",
          "rpm": "5",
          "rpd": "—",
          "tpm": "30K uncached / 90K total",
          "tpd": "1M (TPH 1M)"
        }
      ],
      "card_required": "yes",
      "card_note": "FAQ: credits are granted 'after adding a verified payment method'; without one, Playground and API stay inactive.",
      "trial_credits": "$5 (30 days)",
      "trial_note": null,
      "use_cases": [
        "chat",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "Rate limits + FAQ",
          "url": "https://inference-docs.cerebras.ai/support/rate-limits"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [],
      "notes": [
        "FAQ: 'Is there a permanently free tier? No.'",
        "Token-bucket rate limiting; cached tokens don't count toward the uncached TPM bucket."
      ],
      "best_free_model": "gpt-oss-120b",
      "headline": "$5 trial · 30 days"
    },
    {
      "slug": "sambanova",
      "name": "SambaNova Cloud",
      "category": "LLM inference",
      "website": "https://cloud.sambanova.ai",
      "status": "free-tier",
      "free_type": "Recurring free tier",
      "summary": "Free Tier per model: 20 requests/min, 20 requests/day, 200K tokens/day on DeepSeek, Llama 3.3 70B, gpt-oss-120b, Gemma 4.",
      "free_models": [
        "DeepSeek-V3.1",
        "Meta-Llama-3.3-70B-Instruct",
        "gpt-oss-120b",
        "DeepSeek-V3.2 (preview)",
        "gemma-4-31B-it (preview)"
      ],
      "free_quota": "Per model: 20 RPD and 200,000 tokens/day.",
      "rate_limits": {
        "example_model": "gpt-oss-120b",
        "rpm": 20,
        "rpd": 20,
        "tpm": null,
        "tpd": 200000,
        "other": null
      },
      "limits_table": [
        {
          "model": "DeepSeek-V3.1",
          "rpm": "20",
          "rpd": "20",
          "tpm": "—",
          "tpd": "200K"
        },
        {
          "model": "Meta-Llama-3.3-70B-Instruct",
          "rpm": "20",
          "rpd": "20",
          "tpm": "—",
          "tpd": "200K"
        },
        {
          "model": "gpt-oss-120b",
          "rpm": "20",
          "rpd": "20",
          "tpm": "—",
          "tpd": "200K"
        },
        {
          "model": "DeepSeek-V3.2 (preview)",
          "rpm": "20",
          "rpd": "20",
          "tpm": "—",
          "tpd": "200K"
        },
        {
          "model": "gemma-4-31B-it (preview)",
          "rpm": "20",
          "rpd": "20",
          "tpm": "—",
          "tpd": "200K"
        }
      ],
      "card_required": "unverified",
      "card_note": "Conflicting official pages: the rate-limits doc says the Free Tier 'applies when there is no payment method linked', while the Plans page says 'Add a payment method and purchase credits to run your first requests'.",
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "chat",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "Rate limits policy",
          "url": "https://docs.sambanova.ai/docs/en/models/rate-limits"
        },
        {
          "label": "Plans",
          "url": "https://cloud.sambanova.ai/plans"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "card_required",
        "rate_limits.tpm"
      ],
      "notes": [
        "Developer Tier (payment method linked) is capped at 20M tokens/day across all models."
      ],
      "best_free_model": "DeepSeek-V3.1",
      "headline": "20 req/day per model"
    },
    {
      "slug": "together-ai",
      "name": "Together AI",
      "category": "LLM inference",
      "website": "https://api.together.ai",
      "status": "no-free-tier",
      "free_type": "No free trial",
      "summary": "No free trial: fully prepaid with a $15 minimum first credit purchase.",
      "free_models": [],
      "free_quota": null,
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Dynamic, demand-based limiting (429/503) rather than fixed published numbers."
      },
      "card_required": "yes",
      "card_note": "Minimum $15 credit purchase to access the platform.",
      "trial_credits": null,
      "trial_note": "Docs: 'Together AI does not currently offer free trials.'",
      "use_cases": [],
      "sources": [
        {
          "label": "Billing & credits",
          "url": "https://docs.together.ai/docs/billing"
        },
        {
          "label": "Rate limits",
          "url": "https://docs.together.ai/docs/rate-limits"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [],
      "notes": [
        "After the first $15, minimum top-up is $5.",
        "API access is suspended when balance reaches zero."
      ],
      "best_free_model": null,
      "headline": "No free trial · $15 minimum"
    },
    {
      "slug": "nvidia-nim",
      "name": "NVIDIA NIM (build.nvidia.com)",
      "category": "LLM + multimodal inference",
      "website": "https://build.nvidia.com",
      "status": "free-tier",
      "free_type": "Free prototyping access",
      "summary": "Free NIM API endpoints for prototyping via the NVIDIA Developer Program; production use needs an AI Enterprise license.",
      "free_models": [
        "API catalog models on build.nvidia.com"
      ],
      "free_quota": "Free for prototyping, research, development and testing only.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Per-model limits shown in your build.nvidia.com account; vary by model and current traffic."
      },
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": "Separate 90-day NVIDIA AI Enterprise trial for production-grade self-hosting.",
      "use_cases": [
        "chat",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "NIM FAQ",
          "url": "https://docs.api.nvidia.com/nim/docs/product"
        },
        {
          "label": "build.nvidia.com help",
          "url": "https://build.nvidia.com/search"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits.rpm",
        "card_required"
      ],
      "notes": [
        "build.nvidia.com's help FAQ (as indexed by search engines) states 'up to 40 requests per minute (RPM) for most models'; the page is client-rendered and we could not read it directly, so the RPM field stays blank until verified.",
        "Production use requires an NVIDIA AI Enterprise license (from $4,500/GPU/year per FAQ)."
      ],
      "best_free_model": "API catalog models",
      "headline": "Free for prototyping"
    },
    {
      "slug": "fireworks-ai",
      "name": "Fireworks AI",
      "category": "LLM inference",
      "website": "https://fireworks.ai",
      "status": "trial-only",
      "free_type": "One-time starter credit",
      "summary": "$1 starter credit usable without a payment method, capped at 10 requests/min account-wide.",
      "free_models": [
        "Serverless catalog"
      ],
      "free_quota": "$1 one-time starter credit.",
      "rate_limits": {
        "example_model": "account-wide",
        "rpm": 10,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "10 RPM with no payment method or no credits; 6,000 RPM max with payment method + credits."
      },
      "card_required": "no",
      "card_note": "Without a payment method the account is suspended once the $1 is used.",
      "trial_credits": "$1",
      "trial_note": null,
      "use_cases": [
        "chat",
        "no-card"
      ],
      "sources": [
        {
          "label": "Account quotas",
          "url": "https://docs.fireworks.ai/guides/quotas_usage/account-quotas"
        },
        {
          "label": "How do credits work?",
          "url": "https://docs.fireworks.ai/faq-new/billing-pricing/what-happens-when-i-finish-my-1-dollar-credit"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits.tpm"
      ],
      "notes": [
        "Prepaid credits billing system."
      ],
      "best_free_model": null,
      "headline": "$1 credit · 10 req/min"
    },
    {
      "slug": "deepinfra",
      "name": "DeepInfra",
      "category": "LLM inference",
      "website": "https://deepinfra.com",
      "status": "no-free-tier",
      "free_type": "No free tier",
      "summary": "No sign-up credit: 'You have to add a card or pre-pay or you won't be able to use our services.'",
      "free_models": [],
      "free_quota": null,
      "rate_limits": {
        "example_model": "per model",
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "200 concurrent requests per model (not RPM-based)."
      },
      "card_required": "yes",
      "card_note": "Card or prepayment required (pricing page).",
      "trial_credits": null,
      "trial_note": "DeepStart startup program grants free tokens to approved startups (application only; funding criteria apply).",
      "use_cases": [],
      "sources": [
        {
          "label": "Pricing",
          "url": "https://deepinfra.com/pricing"
        },
        {
          "label": "Rate limits",
          "url": "https://deepinfra.com/docs/advanced/rate-limits"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [],
      "notes": [
        "Several third-party lists still claim '$5 free credits' for DeepInfra; the official pricing page says otherwise."
      ],
      "best_free_model": null,
      "headline": "No free credits"
    },
    {
      "slug": "scaleway",
      "name": "Scaleway Generative APIs",
      "category": "LLM inference (EU)",
      "website": "https://www.scaleway.com/en/generative-apis/",
      "status": "free-tier",
      "free_type": "One-time free tokens",
      "summary": "First 1,000,000 tokens and 60 minutes of audio transcription free; EU-hosted, payment method required.",
      "free_models": [
        "gpt-oss-120b",
        "qwen3.6-35b-a3b",
        "llama-3.3-70b-instruct",
        "mistral-small-3.2-24b-instruct-2506",
        "gemma-4-26b-a4b-it",
        "glm-5.2",
        "qwen3-coder-30b-a3b-instruct",
        "whisper-large-v3"
      ],
      "free_quota": "First 1,000,000 tokens + 60 min audio transcription; charged from token 1,000,001.",
      "rate_limits": {
        "example_model": "most models (payment method validated)",
        "rpm": 300,
        "rpd": null,
        "tpm": 200000,
        "tpd": null,
        "other": "Identity-verified orgs: 600 RPM and 400K–2M TPM depending on model."
      },
      "card_required": "yes",
      "card_note": "Base limits apply once a valid payment method is registered.",
      "trial_credits": "1M tokens + 60 audio min",
      "trial_note": null,
      "use_cases": [
        "chat",
        "coding-agent",
        "embeddings",
        "speech"
      ],
      "sources": [
        {
          "label": "Generative APIs pricing",
          "url": "https://www.scaleway.com/en/pricing/model-as-a-service/"
        },
        {
          "label": "Organization quotas",
          "url": "https://www.scaleway.com/en/docs/organizations-and-projects/additional-content/organization-quotas/"
        },
        {
          "label": "Rate limits",
          "url": "https://www.scaleway.com/en/docs/generative-apis/reference-content/rate-limits/"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [],
      "notes": [
        "Batches API has no rate limit and is billed at -50%."
      ],
      "best_free_model": "gpt-oss-120b",
      "headline": "1M tokens, one-time"
    },
    {
      "slug": "vercel-ai-gateway",
      "name": "Vercel AI Gateway",
      "category": "Gateway / router",
      "website": "https://vercel.com/ai-gateway",
      "status": "free-tier",
      "free_type": "Recurring monthly credit",
      "summary": "$5/month of AI Gateway credit on every Vercel team, for a subset of 'Free Tier eligible' models.",
      "free_models": [
        "Free Tier eligible models (subset of catalog)"
      ],
      "free_quota": "$5 credit per month; starts at your first AI Gateway request.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "'Lower limit per model' on free tier; Vercel does not publish fixed numbers."
      },
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": "Buying any credits moves the team to the paid tier and the $5 monthly free credit no longer applies.",
      "use_cases": [
        "chat",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "AI Gateway pricing",
          "url": "https://vercel.com/docs/ai-gateway/pricing"
        },
        {
          "label": "AI Gateway rate limits",
          "url": "https://vercel.com/docs/ai-gateway/rate-limits"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "card_required",
        "rate_limits"
      ],
      "notes": [
        "BYOK is not available on the free tier.",
        "Zero markup on provider list prices."
      ],
      "best_free_model": "Free-tier models",
      "headline": "$5 credit/month"
    },
    {
      "slug": "ollama-cloud",
      "name": "Ollama Cloud",
      "category": "LLM inference",
      "website": "https://ollama.com",
      "status": "free-tier",
      "free_type": "Recurring starter usage",
      "summary": "Free plan includes a starter amount of cloud usage on starter models, 1 concurrent request; local models are always unlimited.",
      "free_models": [
        "Starter cloud models"
      ],
      "free_quota": "Starter usage credits, reset monthly from sign-up date (amount not published).",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "1 concurrent request on Free (Pro 3, Max/Team 10)."
      },
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "chat",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "Pricing",
          "url": "https://ollama.com/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "free_quota",
        "card_required"
      ],
      "notes": [
        "Paid plans (Pro, Max, Team) include a set dollar amount of usage credits each month; buying usage credits unlocks all models."
      ],
      "best_free_model": "Starter models",
      "headline": "Starter usage/month"
    },
    {
      "slug": "voyage-ai",
      "name": "Voyage AI",
      "category": "Embeddings + rerank",
      "website": "https://www.voyageai.com",
      "status": "free-tier",
      "free_type": "One-time free tokens per model",
      "summary": "First 200M tokens free per account on voyage-4 family & rerank-3; 50M on older domain models.",
      "free_models": [
        "voyage-4-large",
        "voyage-4",
        "voyage-4-lite",
        "voyage-context-4",
        "voyage-code-3",
        "voyage-multimodal-3.5 (200M tokens + 150B pixels)",
        "rerank-3"
      ],
      "free_quota": "200M tokens per listed model (50M for voyage-multilingual-2, -finance-2, -law-2, -code-2).",
      "rate_limits": {
        "example_model": "voyage-4 (Tier 1)",
        "rpm": 2000,
        "rpd": null,
        "tpm": 8000000,
        "tpd": null,
        "other": "Tier 1 requires a payment method; limits without one are not published."
      },
      "card_required": "unverified",
      "card_note": "Free tokens still apply after adding a payment method.",
      "trial_credits": "200M tokens/model",
      "trial_note": null,
      "use_cases": [
        "embeddings",
        "rerank"
      ],
      "sources": [
        {
          "label": "Pricing",
          "url": "https://docs.voyageai.com/docs/pricing"
        },
        {
          "label": "Rate limits",
          "url": "https://docs.voyageai.com/docs/rate-limits"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "card_required"
      ],
      "notes": [
        "Free tokens don't apply to the Batch API.",
        "Older models (e.g. voyage-context-3) have no free tokens."
      ],
      "best_free_model": "voyage-4",
      "headline": "200M tokens per model"
    },
    {
      "slug": "jina-ai",
      "name": "Jina AI (Reader, Embeddings, Rerank)",
      "category": "Embeddings + web reader",
      "website": "https://jina.ai",
      "status": "free-tier",
      "free_type": "Free API key",
      "summary": "Reader works without a key (20 RPM); a free API key raises Reader to 500 RPM and unlocks Embeddings/Rerank at 100 RPM.",
      "free_models": [
        "r.jina.ai Reader",
        "s.jina.ai Search",
        "Embeddings API",
        "Reranker API"
      ],
      "free_quota": null,
      "rate_limits": {
        "example_model": "Embeddings (free key)",
        "rpm": 100,
        "rpd": null,
        "tpm": 100000,
        "tpd": null,
        "other": "Reader: 20 RPM w/o key, 500 RPM with free key."
      },
      "limits_table": [
        {
          "model": "Reader r.jina.ai (no key)",
          "rpm": "20",
          "rpd": "—",
          "tpm": "—",
          "tpd": "—"
        },
        {
          "model": "Reader r.jina.ai (free key)",
          "rpm": "500",
          "rpd": "—",
          "tpm": "—",
          "tpd": "—"
        },
        {
          "model": "Embeddings (free key)",
          "rpm": "100",
          "rpd": "—",
          "tpm": "100K",
          "tpd": "—"
        },
        {
          "model": "Reranker (free key)",
          "rpm": "100",
          "rpd": "—",
          "tpm": "100K",
          "tpd": "—"
        }
      ],
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": "Free-key token allowance is not shown on the public rate-limit page.",
      "use_cases": [
        "embeddings",
        "rerank"
      ],
      "sources": [
        {
          "label": "Rate limit table",
          "url": "https://jina.ai/api-dashboard/rate-limit"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "free_quota",
        "card_required"
      ],
      "notes": [
        "Limits are tracked per API key (or per IP without a key)."
      ],
      "best_free_model": "Embeddings, Reader",
      "headline": "100 req/min (free key)"
    },
    {
      "slug": "z-ai",
      "name": "Z.ai (GLM)",
      "category": "LLM inference",
      "website": "https://z.ai",
      "status": "free-tier",
      "free_type": "Free models",
      "summary": "GLM-4.7-Flash, GLM-4.5-Flash and GLM-4.6V-Flash are priced 'Free' for input and output.",
      "free_models": [
        "GLM-4.7-Flash",
        "GLM-4.5-Flash",
        "GLM-4.6V-Flash (vision)"
      ],
      "free_quota": "Free per-token price on the listed Flash models.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": null
      },
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": "Cached-input storage is 'Limited-time Free' on most paid GLM models.",
      "use_cases": [
        "chat",
        "coding-agent"
      ],
      "sources": [
        {
          "label": "Pricing",
          "url": "https://docs.z.ai/guides/overview/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits",
        "card_required"
      ],
      "notes": [],
      "best_free_model": "GLM-4.7-Flash",
      "headline": "Free per-token price"
    },
    {
      "slug": "assemblyai",
      "name": "AssemblyAI",
      "category": "Speech-to-text",
      "website": "https://www.assemblyai.com",
      "status": "free-tier",
      "free_type": "Free tier (usage-capped)",
      "summary": "No credit card: up to 185 hours of pre-recorded and 333 hours of streaming transcription free.",
      "free_models": [
        "Universal-3.5 Pro",
        "Universal-2",
        "Universal-Streaming"
      ],
      "free_quota": "Up to 185 h pre-recorded + up to 333 h streaming transcription.",
      "rate_limits": {
        "example_model": "Streaming",
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": "Free plan: up to 5 new streaming connections per minute."
      },
      "card_required": "no",
      "card_note": "Pricing FAQ: 'no credit card required'.",
      "trial_credits": null,
      "trial_note": null,
      "use_cases": [
        "speech",
        "no-card"
      ],
      "sources": [
        {
          "label": "Pricing + FAQ",
          "url": "https://www.assemblyai.com/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [],
      "notes": [],
      "best_free_model": "Universal-3.5 Pro",
      "headline": "185 h transcription"
    },
    {
      "slug": "deepgram",
      "name": "Deepgram",
      "category": "Speech-to-text / TTS",
      "website": "https://deepgram.com",
      "status": "trial-only",
      "free_type": "One-time trial credit",
      "summary": "$200 free credit, then pay-as-you-go; no credit card required.",
      "free_models": [
        "Speech-to-text",
        "Text-to-speech",
        "Voice agent"
      ],
      "free_quota": "$200 credit.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": null
      },
      "card_required": "no",
      "card_note": "Pricing page: 'No credit card required.'",
      "trial_credits": "$200",
      "trial_note": null,
      "use_cases": [
        "speech",
        "no-card"
      ],
      "sources": [
        {
          "label": "Pricing",
          "url": "https://deepgram.com/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "rate_limits"
      ],
      "notes": [],
      "best_free_model": "Speech-to-text",
      "headline": "$200 credit"
    },
    {
      "slug": "elevenlabs",
      "name": "ElevenLabs",
      "category": "Text-to-speech / audio",
      "website": "https://elevenlabs.io",
      "status": "free-tier",
      "free_type": "Recurring monthly credits",
      "summary": "Free plan: 10k credits per month across TTS, STT, sound effects and voice design.",
      "free_models": [
        "Text to Speech",
        "Speech to Text",
        "Sound Effects",
        "Voice Design"
      ],
      "free_quota": "10,000 credits per month.",
      "rate_limits": {
        "example_model": null,
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "tpd": null,
        "other": null
      },
      "card_required": "unverified",
      "card_note": null,
      "trial_credits": null,
      "trial_note": "Commercial license starts on Starter ($6/mo per pricing page).",
      "use_cases": [
        "speech"
      ],
      "sources": [
        {
          "label": "Pricing",
          "url": "https://elevenlabs.io/pricing"
        }
      ],
      "last_verified": "2026-10-21",
      "unverified": [
        "card_required",
        "rate_limits"
      ],
      "notes": [],
      "best_free_model": "Text to Speech",
      "headline": "10k credits/month"
    }
  ]
}