[
  {
    "id": "gemini",
    "name": "Google Gemini API",
    "category": "provider-free-tier",
    "signup_url": "https://aistudio.google.com/apikey",
    "base_url": "https://generativelanguage.googleapis.com/v1beta/openai/",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight echoes the requesting origin and allows an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "Free-tier eligibility varies by model",
      "gemini-2.5-flash",
      "gemini-2.5-flash-lite"
    ],
    "limits": {
      "status": "tier-based",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Rate limits apply per project, not per API key, and depend on the project's usage tier; the Free tier is the entry tier and moving up requires enabling billing. Google publishes the tier structure but directs you to AI Studio for the exact RPM, TPM and RPD active on your project, so no single free-tier number can be quoted here.",
      "summary_zh": "限流按项目（project）而不是按 API key 计算，取决于该项目所处的用量层级；Free 是入门层级，往上走需要开启计费。Google 公布了层级结构，但把具体生效的 RPM、TPM 和 RPD 指向 AI Studio 查看，因此这里给不出单一的免费额度数字。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "New projects start on the Free tier. Only selected models carry free-tier usage, and per-model limits are visible in AI Studio.",
      "note_zh": "新项目默认处于 Free 层级。只有部分模型带免费用量，按模型的具体限额在 AI Studio 里可以看到。"
    },
    "official_sources": [
      {
        "title": "Gemini API rate limits",
        "url": "https://ai.google.dev/gemini-api/docs/rate-limits"
      },
      {
        "title": "Gemini API billing",
        "url": "https://ai.google.dev/gemini-api/docs/billing"
      },
      {
        "title": "Gemini OpenAI compatibility",
        "url": "https://ai.google.dev/gemini-api/docs/openai"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "groq",
    "name": "GroqCloud",
    "category": "provider-free-tier",
    "signup_url": "https://console.groq.com/keys",
    "base_url": "https://api.groq.com/openai/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "openai/gpt-oss-120b",
      "openai/gpt-oss-20b",
      "openai/gpt-oss-safeguard-20b",
      "qwen/qwen3.6-27b",
      "groq/compound",
      "groq/compound-mini"
    ],
    "limits": {
      "status": "documented-per-model",
      "requests_per_minute": 30,
      "requests_per_day": 1000,
      "summary": "Free plan limits are per model and enforced per organization. openai/gpt-oss-120b, openai/gpt-oss-20b, openai/gpt-oss-safeguard-20b and qwen/qwen3.6-27b: 30 RPM, 1K RPD, 8K TPM, 200K TPD. groq/compound and groq/compound-mini: 30 RPM, 250 RPD, 70K TPM, no daily token cap listed. meta-llama/llama-prompt-guard-2-22m and -86m: 30 RPM, 14.4K RPD, 15K TPM, 500K TPD. Whisper large-v3 and turbo are metered in audio seconds: 20 RPM, 2K RPD, 7.2K ASH, 28.8K ASD. llama-3.3-70b-versatile and llama-3.1-8b-instant no longer appear in the free table. The RPM/RPD columns here quote openai/gpt-oss-120b.",
      "summary_zh": "免费计划的限额按模型区分，并按组织（organization）合并计算。openai/gpt-oss-120b、openai/gpt-oss-20b、openai/gpt-oss-safeguard-20b、qwen/qwen3.6-27b：30 RPM、1K RPD、8K TPM、200K TPD。groq/compound 和 groq/compound-mini：30 RPM、250 RPD、70K TPM，没有列每日 token 上限。meta-llama/llama-prompt-guard-2-22m 与 -86m：30 RPM、14.4K RPD、15K TPM、500K TPD。Whisper large-v3 与 turbo 按音频秒数计量：20 RPM、2K RPD、7.2K ASH、28.8K ASD。llama-3.3-70b-versatile 和 llama-3.1-8b-instant 已不在免费表内。本页 RPM/RPD 两列取的是 openai/gpt-oss-120b 的数值。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "The Free Plan is available without a payment method; the account Limits page is authoritative for the quota actually active on your organization.",
      "note_zh": "免费计划不需要绑定支付方式；你所在组织实际生效的额度，以账号里的 Limits 页面为准。"
    },
    "official_sources": [
      {
        "title": "GroqCloud rate limits",
        "url": "https://console.groq.com/docs/rate-limits"
      },
      {
        "title": "Groq OpenAI compatibility",
        "url": "https://console.groq.com/docs/openai"
      },
      {
        "title": "Groq billing FAQ",
        "url": "https://console.groq.com/docs/billing-faqs"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "sambanova",
    "name": "SambaNova Cloud",
    "category": "provider-free-tier",
    "signup_url": "https://cloud.sambanova.ai/apis",
    "base_url": "https://api.sambanova.ai/v1",
    "openai_compatible": true,
    "browser_check": "blocked",
    "browser_check_note": "The CORS preflight succeeds but returns no allow-origin header, so a browser refuses to send the Authorization header.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "DeepSeek-V3.1",
      "Meta-Llama-3.3-70B-Instruct",
      "gpt-oss-120b",
      "DeepSeek-V3.2",
      "gemma-4-31B-it"
    ],
    "limits": {
      "status": "documented-per-model",
      "requests_per_minute": 20,
      "requests_per_day": 20,
      "summary": "The Free Tier applies whenever no payment method is linked to the account. Every model in the free table shares the same numbers: 20 RPM, 20 RPD and 200,000 tokens per day. Production models are DeepSeek-V3.1, Meta-Llama-3.3-70B-Instruct and gpt-oss-120b; DeepSeek-V3.2 and gemma-4-31B-it are listed as preview. SambaNova publishes no tokens-per-minute figure for the free tier, only RPM, RPD and TPD. MiniMax-M2.7 appears in the Developer tier tables only. Linking a payment method moves the account to the Developer Tier.",
      "summary_zh": "只要账号没有绑定支付方式，就适用免费档。免费表里每个模型的数字都一样：20 RPM、20 RPD、每天 200,000 token。生产（production）模型是 DeepSeek-V3.1、Meta-Llama-3.3-70B-Instruct 和 gpt-oss-120b；DeepSeek-V3.2 和 gemma-4-31B-it 标着预览（preview）。SambaNova 没有为免费档公布每分钟 token 数，只有 RPM、RPD 和 TPD 三项。MiniMax-M2.7 只出现在开发者档的表里。绑定支付方式后账号转入开发者档。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Free Tier is defined by the absence of a linked payment method, so it is reachable without a card. RPM and RPD are both 20, so one busy minute can spend the whole day.",
      "note_zh": "免费档是以「没绑支付方式」来定义的，所以不用卡就能用到。RPM 和 RPD 都是 20，也就是说忙碌的一分钟就能把一整天的额度花光。"
    },
    "official_sources": [
      {
        "title": "SambaNova model rate limits",
        "url": "https://sambanova-systems.mintlify.dev/docs/en/models/rate-limits.md"
      },
      {
        "title": "SambaNova developer guide",
        "url": "https://docs.sambanova.ai/docs/en/get-started/overview"
      },
      {
        "title": "SambaNova OpenAI compatibility",
        "url": "https://docs.sambanova.ai/docs/en/features/openai-compatibility"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "cohere",
    "name": "Cohere",
    "category": "provider-free-tier",
    "signup_url": "https://dashboard.cohere.com/api-keys",
    "base_url": "https://api.cohere.ai/compatibility/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "command-a-03-2025",
      "command-r-plus",
      "embed-v4.0",
      "rerank-v3.5"
    ],
    "limits": {
      "status": "documented-per-endpoint",
      "requests_per_minute": 20,
      "requests_per_day": null,
      "summary": "Trial keys are free, rate limited and not licensed for production. Chat is limited to 20 requests/minute, Rerank to 10/minute, Audio Transcriptions and EmbedJob to 5/minute, Tokenize to 100/minute, and Embed to 2,000 inputs/minute. Every trial key is additionally capped at 1,000 API calls per month across the account.",
      "summary_zh": "试用 key 免费、有限流，并且按条款不得用于生产。Chat 限 20 次/分钟，Rerank 限 10 次/分钟，Audio Transcriptions 与 EmbedJob 限 5 次/分钟，Tokenize 限 100 次/分钟，Embed 限 2,000 条输入/分钟。此外每个试用 key 在账号维度还有每月 1,000 次 API 调用的封顶。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Trial keys are explicitly excluded from production use by Cohere's terms; the 1,000 calls/month ceiling is the binding limit.",
      "note_zh": "Cohere 的条款明确把试用 key 排除在生产用途之外；每月 1,000 次调用才是真正卡住你的那条线。"
    },
    "official_sources": [
      {
        "title": "Cohere API rate limits",
        "url": "https://docs.cohere.com/docs/rate-limits"
      },
      {
        "title": "Cohere Compatibility API",
        "url": "https://docs.cohere.com/docs/compatibility-api"
      },
      {
        "title": "Cohere pricing",
        "url": "https://cohere.com/pricing"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "cloudflare-workers-ai",
    "name": "Cloudflare Workers AI",
    "category": "provider-free-tier",
    "signup_url": "https://dash.cloudflare.com/profile/api-tokens",
    "base_url": "https://api.cloudflare.com/client/v4/accounts/ACCOUNT_ID/ai/v1",
    "openai_compatible": true,
    "browser_check": "blocked",
    "browser_check_note": "The endpoint answers the CORS preflight with 405 and no allow-origin header, so a browser refuses to send the Authorization header.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "@cf/meta/llama-3.3-70b-instruct-fp8-fast",
      "@cf/openai/gpt-oss-120b",
      "@cf/qwen/qwen2.5-coder-32b-instruct"
    ],
    "limits": {
      "status": "documented-in-compute-units",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Both the Workers Free and Workers Paid plans include 10,000 Neurons per day at no charge, resetting daily at 00:00 UTC. Neurons are a compute unit rather than a request count, so the number of requests you get depends on the model and prompt size. On the Free plan there is no overage: once the allocation is spent, requests fail until reset. Workers Paid bills overage at $0.011 per 1,000 Neurons. A handful of models sit outside the free allowance entirely and require a paid billing method regardless — @cf/moonshotai/kimi-k2.6, @cf/zai-org/glm-5.2 and @cf/deepseek-ai/deepseek-v4-pro-0813 — reachable through Workers Paid or prepaid AI Gateway credits.",
      "summary_zh": "Workers Free 和 Workers Paid 两种计划都含每天 10,000 Neurons 的免费额度，每天 00:00 UTC 重置。Neuron 是计算单位而不是请求次数，所以能换到多少次请求取决于模型和 prompt 大小。免费计划没有超额选项：额度用完，请求就失败到下次重置为止。Workers Paid 的超额价格是每 1,000 Neurons $0.011。有几个模型完全不在免费额度覆盖范围内，无论如何都要先有付费方式 —— @cf/moonshotai/kimi-k2.6、@cf/zai-org/glm-5.2 和 @cf/deepseek-ai/deepseek-v4-pro-0813 —— 要走 Workers Paid 或预付的 AI Gateway 额度。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "The pricing page does not state whether a credit card is required to sign up for the Workers Free plan, so treat the card column as unverified for this entry.",
      "note_zh": "定价页没有说明注册 Workers Free 计划是否需要信用卡，所以这一条的\"需信用卡\"按未核实处理。"
    },
    "official_sources": [
      {
        "title": "Workers AI pricing",
        "url": "https://developers.cloudflare.com/workers-ai/platform/pricing/"
      },
      {
        "title": "Workers AI OpenAI compatible endpoints",
        "url": "https://developers.cloudflare.com/workers-ai/configuration/open-ai-compatibility/"
      },
      {
        "title": "Workers AI models",
        "url": "https://developers.cloudflare.com/workers-ai/models/"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "huggingface",
    "name": "Hugging Face Inference Providers",
    "category": "provider-free-tier",
    "signup_url": "https://huggingface.co/settings/tokens",
    "base_url": "https://router.huggingface.co/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "deepseek-ai/DeepSeek-V3-0324",
      "openai/gpt-oss-120b",
      "200+ models routed across partner providers"
    ],
    "limits": {
      "status": "documented-in-credits",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Signed-in free users receive $0.10 in monthly Inference Provider credits (Hugging Face notes this is subject to change); PRO users receive $2.00 and Team or Enterprise organizations $2.00 per seat. Credits only apply to requests routed by Hugging Face, not to requests made with your own provider key. Once the monthly credits are spent you must purchase credits to continue.",
      "summary_zh": "登录的免费用户每月获得 0.10 美元的 Inference Provider 额度（Hugging Face 注明这个数额可能调整）；PRO 用户 2.00 美元，Team 或 Enterprise 组织按每席位 2.00 美元。额度只作用于经 Hugging Face 路由的请求，用你自己的服务商 key 直连既不消耗也不享受。每月额度花完后需要另行购买才能继续。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Hugging Face charges provider list rates with no markup, so the free allowance converts to very different request counts depending on which model you route to.",
      "note_zh": "Hugging Face 按各服务商的原价计费、不加价，所以同样一份额度换算成请求数，会因为路由到哪个模型而差别极大。"
    },
    "official_sources": [
      {
        "title": "Inference Providers pricing and billing",
        "url": "https://huggingface.co/docs/inference-providers/pricing"
      },
      {
        "title": "Inference Providers OpenAI compatibility",
        "url": "https://huggingface.co/docs/inference-providers/index"
      },
      {
        "title": "Inference Providers settings",
        "url": "https://huggingface.co/settings/inference-providers"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "siliconflow",
    "name": "SiliconFlow",
    "category": "provider-free-tier",
    "signup_url": "https://cloud.siliconflow.com/account/ak",
    "base_url": "https://api.siliconflow.com/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "Qwen/Qwen3-8B",
      "THUDM/GLM-4-9B-0414",
      "deepseek-ai/DeepSeek-R1"
    ],
    "limits": {
      "status": "documented-per-tier",
      "requests_per_minute": 1000,
      "requests_per_day": null,
      "summary": "Rate limits for free models are fixed; paid models are tiered by monthly spend and start at tier L0 with 1,000 RPM and 40,000 TPM. Limits are enforced per user account rather than per API key, and each model is limited separately. deepseek-ai/DeepSeek-R1 and deepseek-ai/DeepSeek-V3 carry an extra cap of 30 requests/hour and 100 requests/day.",
      "summary_zh": "免费模型的限流是固定值；付费模型按月消费分层，L0 起步为 1,000 RPM、40,000 TPM。限流按用户账号而不是按 API key 计算，而且每个模型单独计限。deepseek-ai/DeepSeek-R1 与 deepseek-ai/DeepSeek-V3 另有每小时 30 次、每天 100 次的额外封顶。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "The 1,000 RPM figure is the L0 paid-model ceiling. Free models have their own fixed limits published per model on the model pages.",
      "note_zh": "1,000 RPM 是付费模型 L0 档的上限。免费模型有各自固定的限额，分别公布在对应的模型页上。"
    },
    "official_sources": [
      {
        "title": "SiliconFlow rate limits",
        "url": "https://docs.siliconflow.com/en/userguide/rate-limits/rate-limit-and-upgradation"
      },
      {
        "title": "SiliconFlow quick start",
        "url": "https://docs.siliconflow.com/en/userguide/quickstart"
      },
      {
        "title": "SiliconFlow models",
        "url": "https://cloud.siliconflow.com/models"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "fireworks",
    "name": "Fireworks AI",
    "category": "provider-free-tier",
    "signup_url": "https://app.fireworks.ai/settings/users/api-keys",
    "base_url": "https://api.fireworks.ai/inference/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "accounts/fireworks/models/llama-v3p3-70b-instruct",
      "accounts/fireworks/models/gpt-oss-120b"
    ],
    "limits": {
      "status": "documented-account-wide",
      "requests_per_minute": 10,
      "requests_per_day": null,
      "summary": "An account with no payment method and no credits is limited to 10 requests per minute across the entire account. Adding a payment method and active credits raises the ceiling to a maximum of 6,000 RPM. Fireworks is pre-paid, so the 10 RPM envelope is the only no-cost path and there is no published daily request cap on it.",
      "summary_zh": "既没有绑定支付方式、也没有余额的账号，整个账号合计限 10 次请求/分钟。绑定支付方式并有可用余额后，上限最高能到 6,000 RPM。Fireworks 是预付费模式，所以 10 RPM 这个区间是唯一的零成本路径，官方也没有为它公布每日请求上限。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "The 10 RPM limit is a request-rate envelope, not a free token grant; serverless token throughput still depends on your spending tier.",
      "note_zh": "10 RPM 是请求速率的封顶，不是一份免费 token 赠额；serverless 的 token 吞吐仍然取决于你的消费层级。"
    },
    "official_sources": [
      {
        "title": "Fireworks account quotas",
        "url": "https://docs.fireworks.ai/guides/quotas_usage/account-quotas"
      },
      {
        "title": "Fireworks serverless rate limits",
        "url": "https://docs.fireworks.ai/guides/quotas_usage/rate-limits"
      },
      {
        "title": "Fireworks pricing",
        "url": "https://fireworks.ai/pricing"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "zai",
    "name": "Z.AI Open Platform",
    "category": "provider-free-tier",
    "signup_url": "https://z.ai/manage-apikey/apikey-list",
    "base_url": "https://api.z.ai/api/paas/v4",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight echoes the requesting origin and allows an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "GLM-4.7-Flash",
      "GLM-4.5-Flash",
      "GLM-4.6V-Flash"
    ],
    "limits": {
      "status": "free-models-listed",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "GLM-4.7-Flash, GLM-4.5-Flash and the vision model GLM-4.6V-Flash are listed at $0 for input, cached input, cached input storage and output on the official pricing table, making them free to call rather than merely discounted. Z.AI does not publish per-model RPM or TPM for these models on the pricing page; the rate limit reference is a separate page and the console is authoritative.",
      "summary_zh": "GLM-4.7-Flash、GLM-4.5-Flash 和视觉模型 GLM-4.6V-Flash 在官方定价表上的输入、缓存输入、缓存输入存储和输出四列都标着 $0，也就是真的免费调用，不是打折。Z.AI 没有在定价页上公布这几个模型的 RPM 或 TPM；限流说明在另一页，最终以控制台为准。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "The free Flash models sit alongside paid GLM models on the same key, so watch which model ID you send to avoid unintended charges.",
      "note_zh": "免费的 Flash 模型和付费的 GLM 模型共用同一个 key，所以要留意发出去的 model ID，免得意外产生费用。"
    },
    "official_sources": [
      {
        "title": "Z.AI pricing",
        "url": "https://docs.z.ai/guides/overview/pricing"
      },
      {
        "title": "Z.AI rate limits",
        "url": "https://docs.z.ai/api-reference/rate-limit"
      },
      {
        "title": "OpenAI Python SDK with Z.AI",
        "url": "https://docs.z.ai/guides/develop/openai/python"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "novita",
    "name": "Novita AI",
    "category": "metered-access",
    "signup_url": "https://novita.ai/settings/key-management",
    "base_url": "https://api.novita.ai/openai",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "Llama 3.1 8B Instruct",
      "OpenAI: GPT OSS 20B",
      "BAAI:BGE-M3"
    ],
    "limits": {
      "status": "not_published",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "No model on the pricing table is priced at zero any more. The two entries this catalog previously listed as Free — inclusionai/Ling-3.0-flash and Mind Lab Macaron V1 Venti — are gone from the table. The cheapest rows are now Llama 3.1 8B Instruct at $0.02 input / $0.05 output per million and BAAI:BGE-M3 embeddings at $0.01 per million. No signup credit is published, and the page states no request-rate numbers.",
      "summary_zh": "定价表里已经没有任何一个模型标价为零。本清单原先列的两个免费项 — inclusionai/Ling-3.0-flash 和 Mind Lab Macaron V1 Venti — 已从表中消失。现在最便宜的是 Llama 3.1 8B Instruct（每百万 token 输入 $0.02、输出 $0.05）和 BAAI:BGE-M3 向量模型（每百万 $0.01）。页面没有公布注册赠额，也没有公布任何请求速率数字。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Moved out of the free-tier list because the free rows disappeared, not because the service did. Batch inference carries an introductory 50% discount.",
      "note_zh": "从免费清单里移出，是因为免费的那几行没了，不是服务没了。批量推理有一个 50% 的入门折扣。"
    },
    "official_sources": [
      {
        "title": "Novita pricing",
        "url": "https://novita.ai/pricing"
      },
      {
        "title": "Novita LLM API reference",
        "url": "https://novita.ai/docs/api-reference/model-apis-llm-create-chat-completion"
      },
      {
        "title": "Novita model library",
        "url": "https://novita.ai/models"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "mistral",
    "name": "Mistral La Plateforme",
    "category": "provider-free-tier",
    "signup_url": "https://console.mistral.ai/api-keys",
    "base_url": "https://api.mistral.ai/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "mistral-small-latest",
      "open-mistral-nemo",
      "codestral-latest"
    ],
    "limits": {
      "status": "not_published",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Mistral documents a Free mode (also referred to as the Experiment plan) that lets you activate a workspace and generate an API key without paying, but the public documentation does not currently expose a reachable rate-limit page for it: the tier URL listed in Mistral's own documentation index returns 404 as of the check date. The console is the only authoritative source for the Experiment plan's requests-per-second and tokens-per-minute ceilings.",
      "summary_zh": "Mistral 文档里有一个 Free 模式（也称 Experiment 计划），不付费就能激活工作区并生成 API key，但公开文档目前没有一个打得开的限流页面：核验当天，Mistral 自己文档索引里给出的该层级 URL 返回 404。Experiment 计划的每秒请求数和每分钟 token 数，只能以控制台为准。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Mistral's consumer Free plan for the Vibe assistant is a different product from the API Free mode; do not read the assistant's message limits as API limits.",
      "note_zh": "Mistral 面向消费者的 Vibe 助手也有个 Free 计划，那是另一个产品；不要把助手的消息条数读成 API 的限额。"
    },
    "official_sources": [
      {
        "title": "Mistral La Plateforme documentation",
        "url": "https://docs.mistral.ai/"
      },
      {
        "title": "Mistral pricing",
        "url": "https://mistral.ai/pricing"
      },
      {
        "title": "Mistral API documentation index",
        "url": "https://docs.mistral.ai/llms.txt"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "dashscope",
    "name": "Alibaba Cloud Model Studio",
    "category": "provider-free-tier",
    "signup_url": "https://bailian.console.alibabacloud.com/",
    "base_url": "https://dashscope-intl.aliyuncs.com/compatible-mode/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "qwen-plus",
      "qwen-turbo",
      "qwen3-coder-plus"
    ],
    "limits": {
      "status": "documented-per-model",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Rate limiting is applied at the Alibaba Cloud root account level and aggregates usage across all RAM users, workspaces and API keys under that account. Each model carries its own RPM and TPM limit published in per-model tables, and free quota is granted per model rather than as one account-wide allowance, so no single free-tier number applies.",
      "summary_zh": "限流在阿里云主账号这一层生效，会把该账号下所有 RAM 用户、工作空间和 API key 的用量合并计算。每个模型有各自的 RPM 和 TPM，公布在按模型的表格里，免费额度也是按模型发放而不是给一份账号级总额，因此没有单一的免费额度数字。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Requests are rejected when a limit is hit and typically recover within one minute; the error text tells you whether RPM or TPM triggered.",
      "note_zh": "触发限流时请求会被拒绝，通常一分钟内恢复；报错文本会告诉你触发的是 RPM 还是 TPM。"
    },
    "official_sources": [
      {
        "title": "Model Studio rate limiting",
        "url": "https://www.alibabacloud.com/help/en/model-studio/rate-limit"
      },
      {
        "title": "Model Studio OpenAI compatibility",
        "url": "https://www.alibabacloud.com/help/en/model-studio/compatibility-of-openai-with-dashscope"
      },
      {
        "title": "Model Studio models",
        "url": "https://www.alibabacloud.com/help/en/model-studio/models"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "moonshot",
    "name": "Moonshot AI (Kimi)",
    "category": "metered-access",
    "signup_url": "https://platform.moonshot.ai/console/api-keys",
    "base_url": "https://api.moonshot.ai/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight echoes the requesting origin and allows an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "kimi-k2-0905-preview",
      "moonshot-v1-8k",
      "moonshot-v1-128k"
    ],
    "limits": {
      "status": "documented-per-tier",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "The tier table is keyed to cumulative recharge rather than to a standing free allowance: the docs state you must recharge at least $1 before you can start using the platform, and Tier0 then gives 1 concurrent request, 3 RPM, 500K TPM and 1.5M TPD. Tier1 at $10 cumulative gives 50 concurrent, 200 RPM, 2M TPM and unlimited TPD, rising to Tier5 at $3,000. A $5 voucher arrives only once cumulative payments reach $5, and vouchers do not count toward that cumulative total. The docs moved to platform.kimi.ai and the old platform.moonshot.ai paths now 301 there; the page also carries a notice that the tier rules were slated for revision in August because of abusive traffic.",
      "summary_zh": "分档表挂钩的是累计充值金额，而不是一份长期免费额度：文档写明必须先充值至少 $1 才能开始使用，此后 Tier0 给 1 路并发、3 RPM、500K TPM、1.5M TPD。累计 $10 的 Tier1 给 50 路并发、200 RPM、2M TPM、TPD 不限，一直到 Tier5 的 $3,000。$5 代金券要等累计付款达到 $5 才发放，而代金券本身不计入累计充值。文档已迁到 platform.kimi.ai，原先 platform.moonshot.ai 的路径现在 301 过去；页面另有一条告示：因异常流量，分档与限额规则拟于 8 月修订。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Not a free tier in practice: the lowest bracket still requires a $1 payment before any call goes through. Kept in the catalog so the $1 floor is visible rather than discovered at signup.",
      "note_zh": "实际上不算免费额度：最低那一档也要先付 $1 才调得通。留在清单里，是为了让这道 $1 的门槛提前看得见，而不是注册到一半才发现。"
    },
    "official_sources": [
      {
        "title": "Kimi rate limits",
        "url": "https://platform.kimi.ai/docs/pricing/limits"
      },
      {
        "title": "Kimi API overview",
        "url": "https://platform.kimi.ai/docs/introduction"
      },
      {
        "title": "Kimi pricing",
        "url": "https://platform.kimi.ai/docs/pricing/chat"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "pollinations",
    "name": "Pollinations.AI",
    "category": "provider-free-tier",
    "signup_url": "https://pollinations.ai/",
    "base_url": "https://text.pollinations.ai/openai",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "openai",
      "mistral",
      "Community-hosted open models"
    ],
    "limits": {
      "status": "documented-per-tier",
      "requests_per_minute": 4,
      "requests_per_day": null,
      "summary": "The API docs now publish four tiers by minimum interval between requests: Anonymous, no signup, basic models, one request every 15 seconds; Seed, free registration, standard models, one request every 5 seconds; Flower, paid, advanced models, one request every 3 seconds; Nectar, enterprise, all models, no limit. The RPM column here is the Anonymous tier expressed per minute (one per 15s = 4/min). Since 2025-03-31 free-tier images may carry a watermark, and removing it with the nologo parameter needs an account.",
      "summary_zh": "API 文档现在按「两次请求的最小间隔」公布了四档：Anonymous，免注册，基础模型，每 15 秒 1 次；Seed，免费注册，标准模型，每 5 秒 1 次；Flower，付费，高级模型，每 3 秒 1 次；Nectar，企业版，全部模型，不限。本页 RPM 一列取的是 Anonymous 档折算成每分钟的值（每 15 秒 1 次 = 每分钟 4 次）。自 2025-03-31 起免费档产出的图片可能带水印，用 nologo 参数去掉水印需要账号。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Still callable with no key at all, which is what makes it useful for a first test; treat availability as best-effort and do not build production dependencies on it.",
      "note_zh": "仍然可以完全不带 key 直接调，这正是它适合拿来做第一次试探的原因；可用性按尽力而为对待，不要让生产环境依赖它。"
    },
    "official_sources": [
      {
        "title": "Pollinations API documentation",
        "url": "https://github.com/pollinations/pollinations/blob/master/APIDOCS.md"
      },
      {
        "title": "Pollinations project repository",
        "url": "https://github.com/pollinations/pollinations"
      },
      {
        "title": "Pollinations.AI home",
        "url": "https://pollinations.ai/"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "ollama-cloud",
    "name": "Ollama Cloud",
    "category": "provider-free-tier",
    "signup_url": "https://ollama.com/settings/keys",
    "base_url": "https://ollama.com/v1",
    "openai_compatible": true,
    "browser_check": "blocked",
    "browser_check_note": "The endpoint answers the CORS preflight with 405 and no allow-origin header, so a browser refuses to send the Authorization header.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "gpt-oss:120b-cloud",
      "gpt-oss:20b-cloud",
      "qwen3-coder:480b-cloud"
    ],
    "limits": {
      "status": "not_published",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Cloud models run on Ollama's servers while keeping the local CLI workflow, and require only an ollama.com account to start. The Cloud documentation page describes access and model retirement policy but does not publish hourly or daily request limits, so the account page is the authoritative source for the active quota.",
      "summary_zh": "云端模型跑在 Ollama 自己的服务器上，本地 CLI 的用法保持不变，只要一个 ollama.com 账号就能开始。Cloud 文档页说明了访问方式和模型下线策略，但没有公布每小时或每天的请求限额，实际生效的额度以账号页面为准。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Ollama periodically retires older cloud models as newer open-weight models ship; local models are unaffected by those retirements.",
      "note_zh": "随着新的开放权重模型发布，Ollama 会定期下线较旧的云端模型；本地模型不受这些下线影响。"
    },
    "official_sources": [
      {
        "title": "Ollama Cloud documentation",
        "url": "https://docs.ollama.com/cloud"
      },
      {
        "title": "Ollama Cloud API access",
        "url": "https://docs.ollama.com/api"
      },
      {
        "title": "Ollama model library",
        "url": "https://ollama.com/library"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "cerebras",
    "name": "Cerebras Inference",
    "category": "trial-credit",
    "signup_url": "https://cloud.cerebras.ai/",
    "base_url": "https://api.cerebras.ai/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": true,
    "models": [
      "gpt-oss-120b",
      "gemma-4-31b"
    ],
    "limits": {
      "status": "documented-per-model",
      "requests_per_minute": 5,
      "requests_per_day": null,
      "summary": "Free Trial limits are 5 RPM and 30K TPM per model, capped at 1M tokens per hour and 1M tokens per day, for gpt-oss-120b and gemma-4-31b — zai-glm-4.7 is no longer in that table. gemma-4-31b additionally caps images at 2 per request and 4 MB per payload. New accounts receive $5 in credits that expire 30 days after being granted, and Cerebras states plainly that it offers no always-free per-model allowance. Skipping the payment method at sign-up leaves Playground and API access inactive.",
      "summary_zh": "免费试用的限额是每个模型 5 RPM、30K TPM，并有每小时 100 万 token、每天 100 万 token 的封顶，适用于 gpt-oss-120b 和 gemma-4-31b —— zai-glm-4.7 已不在这张表里。gemma-4-31b 另有图片限制：每次请求 2 张、单次载荷 4 MB。新账号获得 5 美元额度，自发放起 30 天过期；Cerebras 明确说明不提供按模型的长期免费额度。注册时跳过绑定支付方式，Playground 和 API 访问会一直是未激活状态。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "This entry is marked as requiring a card because a verified payment method is a precondition for API access, even though the trial itself costs nothing.",
      "note_zh": "这一条标为需要信用卡，是因为通过验证的支付方式是拿到 API 访问权的前置条件，尽管试用本身不花钱。"
    },
    "official_sources": [
      {
        "title": "Cerebras rate limits",
        "url": "https://inference-docs.cerebras.ai/support/rate-limits"
      },
      {
        "title": "Cerebras pricing",
        "url": "https://inference-docs.cerebras.ai/support/pricing"
      },
      {
        "title": "Cerebras OpenAI compatibility",
        "url": "https://inference-docs.cerebras.ai/resources/openai"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "vercel-ai-gateway",
    "name": "Vercel AI Gateway",
    "category": "trial-credit",
    "signup_url": "https://vercel.com/dashboard/ai-gateway",
    "base_url": "https://ai-gateway.vercel.sh/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight echoes the requesting origin and allows an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "Free Tier eligible model subset",
      "openai/gpt-oss-120b",
      "moonshotai/kimi-k2"
    ],
    "limits": {
      "status": "documented-in-credits",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Every Vercel team account gets a free tier, but it covers a subset of models rather than the full catalog, and free-tier requests are rate limited per model at lower limits than the paid tier — those per-model numbers are not published. The free credits start counting from your first Gateway request. The pricing page no longer states a dollar figure for the monthly credit: the $5 per month this catalog previously quoted is gone from the page, so no amount is quoted here.",
      "summary_zh": "每个 Vercel team 账号都有免费档，但免费档只覆盖模型子集，不是全部目录；免费档的请求还按模型限流，上限低于付费档，而这些按模型的数字并不公布。免费额度从你第一次调用 Gateway 时开始计。定价页已经不再写出每月免费额度的具体金额：本清单原先引用的每月 5 美元已经不在页面上，所以这里不再给出金额。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Purchasing credits moves the team to the paid tier and the monthly free credit stops applying — it is not additive.",
      "note_zh": "一旦购买额度，team 就转到付费档，每月的免费额度不再生效 —— 两者不叠加。"
    },
    "official_sources": [
      {
        "title": "AI Gateway pricing",
        "url": "https://vercel.com/docs/ai-gateway/pricing"
      },
      {
        "title": "AI Gateway getting started",
        "url": "https://vercel.com/docs/ai-gateway/getting-started"
      },
      {
        "title": "AI Gateway models",
        "url": "https://vercel.com/docs/ai-gateway/models-and-providers"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "watsonx",
    "name": "IBM watsonx.ai",
    "category": "trial-credit",
    "signup_url": "https://dataplatform.cloud.ibm.com/registration/stepone",
    "base_url": "https://us-south.ml.cloud.ibm.com/ml/v1",
    "openai_compatible": false,
    "browser_check": "blocked",
    "browser_check_note": "The CORS preflight succeeds but returns no allow-origin header, so a browser refuses to send the Authorization header.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "ibm/granite-3-8b-instruct",
      "meta-llama/llama-3-3-70b-instruct",
      "mistralai/mistral-large"
    ],
    "limits": {
      "status": "not_published",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "IBM offers a no-cost trial of watsonx.ai alongside the paid Essentials and Standard plans, and the pricing page invites you to start building at no cost. The page states plan pricing per resource unit but does not publish request-rate limits for the trial, so the console is authoritative for the trial's ceilings.",
      "summary_zh": "IBM 在付费的 Essentials 和 Standard 计划之外，提供 watsonx.ai 的免费试用，定价页也明确邀请你零成本开始构建。该页面按资源单位列出了各计划的价格，但没有公布试用的请求速率限制，试用的上限以控制台为准。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "watsonx.ai is not OpenAI-wire-compatible by default; requests use IBM's own ML text-generation API shape.",
      "note_zh": "watsonx.ai 默认不兼容 OpenAI 协议；请求走的是 IBM 自己的 ML 文本生成 API 格式。"
    },
    "official_sources": [
      {
        "title": "watsonx.ai pricing",
        "url": "https://www.ibm.com/products/watsonx-ai/pricing"
      },
      {
        "title": "watsonx.ai API reference",
        "url": "https://cloud.ibm.com/apidocs/watsonx-ai"
      },
      {
        "title": "watsonx.ai foundation models",
        "url": "https://www.ibm.com/products/watsonx-ai/foundation-models"
      }
    ],
    "source_checked_at": "2026-07-25",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "openrouter",
    "name": "OpenRouter",
    "category": "free-model-aggregator",
    "signup_url": "https://openrouter.ai/settings/keys",
    "base_url": "https://openrouter.ai/api/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "Model IDs ending in :free",
      "openrouter/free"
    ],
    "limits": {
      "status": "documented-with-conditions",
      "requests_per_minute": 20,
      "requests_per_day": 50,
      "summary": "Model variants whose ID ends in :free are capped at 20 requests/minute regardless of account status. The daily cap depends on lifetime credit purchases: under 10 credits gives 50 requests/day, and 10 or more credits raises it to 1,000 requests/day. OpenRouter governs capacity globally, so extra accounts or extra API keys do not raise these limits.",
      "summary_zh": "model ID 以 :free 结尾的变体一律限 20 次请求/分钟，与账号状态无关。每日上限取决于历史累计购买的额度：不足 10 credits 是每天 50 次请求，达到 10 credits 及以上提升到每天 1,000 次。OpenRouter 在全局层面调度容量，所以多开账号或多建 API key 都不会把这些限额抬高。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "A negative account balance can return 402 errors even on free models until the balance is topped back above zero.",
      "note_zh": "账号余额为负时，即使调用免费模型也可能收到 402，直到把余额补回零以上为止。"
    },
    "official_sources": [
      {
        "title": "OpenRouter API rate limits",
        "url": "https://openrouter.ai/docs/api-reference/limits"
      },
      {
        "title": "OpenRouter free model variants",
        "url": "https://openrouter.ai/docs/guides/routing/model-variants/free"
      },
      {
        "title": "OpenRouter quickstart",
        "url": "https://openrouter.ai/docs/quickstart"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "github-models",
    "name": "GitHub Models",
    "category": "retired-free-tier",
    "signup_url": null,
    "base_url": "https://models.github.ai/inference",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "Retired — the model catalog is gone"
    ],
    "limits": {
      "status": "retired",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Retired on 2026-07-30. GitHub shut down the playground, the model catalog, the inference API and the BYOK endpoints; the changelog entry dated that day reads “GitHub Models is now retired.” There is no free tier left to quote. GitHub points existing users at Microsoft Foundry for a broad model catalog and at GitHub Copilot for AI workflows.",
      "summary_zh": "2026-07-30 已下线。GitHub 关停了 playground、模型目录、推理 API 和 BYOK 端点；当天的更新日志把它标注为已下线（英文原话见下方来源链接）。已经没有免费额度可以引用。GitHub 让原有用户转向 Microsoft Foundry（模型选择面广）或 GitHub Copilot（AI 工作流）。"
    },
    "availability": {
      "status": "retired",
      "accepting_new_users": false,
      "retires_at": "2026-07-30",
      "note": "Kept in the catalog as a tombstone: the shutdown covered all customers including those with active usage, and searches for a GitHub Models free tier still land here.",
      "note_zh": "留在清单里当一块墓碑：这次关停覆盖所有客户，包括仍在用的；而搜「GitHub Models 免费额度」的人还是会找到这里。"
    },
    "official_sources": [
      {
        "title": "GitHub Models is being fully retired on July 30, 2026",
        "url": "https://github.blog/changelog/2026-07-01-github-models-is-being-fully-retired-on-july-30-2026/"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "together",
    "name": "Together AI",
    "category": "metered-access",
    "signup_url": "https://api.together.ai/settings/api-keys",
    "base_url": "https://api.together.xyz/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "meta-llama/Llama-3.3-70B-Instruct-Turbo",
      "openai/gpt-oss-120b",
      "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo"
    ],
    "limits": {
      "status": "dynamic-no-fixed-numbers",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Together applies dynamic per-model rate limits that rise with sustained successful traffic and fall when traffic drops, and states plainly that there are no fixed per-model limits published. Requests above your dynamic rate return 429 with x-ratelimit-reset; requests at or below it that still fail return 503. No standing free allowance is documented on the rate-limits page.",
      "summary_zh": "Together 采用按模型的动态限流：持续成功的流量会把限额抬上去，流量降下来限额也跟着回落，官方明确说明不公布固定的按模型限额。超过动态速率的请求返回 429 并带 x-ratelimit-reset；在限额之内却仍然失败的请求返回 503。限流页面上没有记录任何常设的免费额度。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Because limits are usage-derived, a new account cannot know its ceiling in advance; read the 429 response headers instead.",
      "note_zh": "因为限额是从用量推导出来的，新账号无法事先知道自己的上限；改读 429 的响应头才是办法。"
    },
    "official_sources": [
      {
        "title": "Together serverless rate limits",
        "url": "https://docs.together.ai/docs/serverless/rate-limits"
      },
      {
        "title": "Together OpenAI compatibility",
        "url": "https://docs.together.ai/docs/openai-api-compatibility"
      },
      {
        "title": "Together pricing",
        "url": "https://www.together.ai/pricing"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "nebius",
    "name": "Nebius Token Factory",
    "category": "metered-access",
    "signup_url": "https://tokenfactory.nebius.com/",
    "base_url": "https://api.tokenfactory.nebius.com/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "deepseek-ai/DeepSeek-V3",
      "meta-llama/Llama-3.3-70B-Instruct",
      "Qwen/Qwen3-235B-A22B"
    ],
    "limits": {
      "status": "dynamic-no-fixed-numbers",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Limits are dynamic and the docs do not publish an account default: they tell you to read the Rate Limits page inside Token Factory. The 60 RPM / 400,000 TPM pair this catalog previously carried as a baseline appears on the page only inside a table labelled a visual example, so it is not quoted as a limit here. The published rules that do apply: usage is evaluated in rolling 15-minute windows, averaging at or above 80% of the current limit raises it by 20% for the next window, averaging at or below 50% divides it by 1.5, and the ceiling is 20x the base allocation before an Enterprise plan is required. No standing free allowance is documented.",
      "summary_zh": "限额是动态的，文档不公布账号默认值：它让你去 Token Factory 里的 Rate Limits 页面自己看。本清单原先当作基准值的 60 RPM / 400,000 TPM，在页面上只出现在一张标着「示意」（visual example）的表里，所以这里不再把它当限额引用。真正通用的规则是：按 15 分钟滚动窗口评估用量，窗口内平均达到当前限额的 80% 及以上，下一窗口上调 20%；平均在 50% 及以下，则除以 1.5；上限是基准配额的 20 倍，再往上要走企业版。文档没有记载任何长期免费额度。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Requests above the limit may still be served at lower priority, flagged with an x-ratelimit-over-limit header, rather than failing outright.",
      "note_zh": "超出限额的请求可能仍然会以较低优先级被处理，并带上 x-ratelimit-over-limit 响应头，而不是直接失败。"
    },
    "official_sources": [
      {
        "title": "Nebius rate limits and scaling",
        "url": "https://docs.tokenfactory.nebius.com/ai-models-inference/rate-limits"
      },
      {
        "title": "Nebius Token Factory quickstart",
        "url": "https://docs.tokenfactory.nebius.com/quickstart"
      },
      {
        "title": "Nebius billing and consumption",
        "url": "https://docs.tokenfactory.nebius.com/other-capabilities/billing-new"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "perplexity",
    "name": "Perplexity API",
    "category": "metered-access",
    "signup_url": "https://www.perplexity.ai/account/api/keys",
    "base_url": "https://api.perplexity.ai",
    "openai_compatible": true,
    "browser_check": "blocked",
    "browser_check_note": "The endpoint answers the CORS preflight with 404 and no allow-origin header, so a browser refuses to send the Authorization header.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "sonar",
      "sonar-pro",
      "sonar-reasoning"
    ],
    "limits": {
      "status": "documented-per-tier",
      "requests_per_minute": 50,
      "requests_per_day": null,
      "summary": "Usage tiers are set by cumulative API credit purchases and never downgrade. Tier 0 ($0 purchased) allows 1 query/second and 50 requests/minute on the Agent API; Tier 1 ($50+) allows 3 QPS and 150/min, rising to 33 QPS and 2,000/min at Tier 4. The Search API is separately limited to 50 requests/second at every tier. Tier 0 sets a rate ceiling but is not a free token grant, so credits are still required to make calls.",
      "summary_zh": "用量层级由累计购买的 API 额度决定，而且只升不降。Tier 0（累计购买 0 美元）在 Agent API 上允许 1 QPS、50 次请求/分钟；Tier 1（50 美元以上）是 3 QPS、150 次/分钟，到 Tier 4 提升到 33 QPS、2,000 次/分钟。Search API 在所有层级都单独限为 50 次请求/秒。Tier 0 只是划定了速率上限，并不是一份免费 token 赠额，调用之前仍然要先有余额。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Perplexity Pro subscribers receive a monthly API credit allowance through their subscription; that allowance is not documented on the rate-limits page.",
      "note_zh": "Perplexity Pro 订阅者会通过订阅拿到每月的 API 额度；这份额度没有写在限流页面上。"
    },
    "official_sources": [
      {
        "title": "Perplexity rate limits and usage tiers",
        "url": "https://docs.perplexity.ai/docs/admin/rate-limits-usage-tiers"
      },
      {
        "title": "Perplexity pricing",
        "url": "https://docs.perplexity.ai/docs/getting-started/pricing"
      },
      {
        "title": "Perplexity OpenAI compatibility",
        "url": "https://docs.perplexity.ai/docs/agent-api/openai-compatibility"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "deepinfra",
    "name": "DeepInfra",
    "category": "metered-access",
    "signup_url": "https://deepinfra.com/dash/api_keys",
    "base_url": "https://api.deepinfra.com/v1/openai",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "deepseek-ai/DeepSeek-V3",
      "Qwen/Qwen3-Next-80B-A3B-Instruct",
      "meta-llama/Llama-4-Scout-17B-16E"
    ],
    "limits": {
      "status": "not_published",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "The official pricing page lists a per-million-token input and output price for every serving model and does not describe a free allowance or publish request-rate limits. Treat DeepInfra as pay-as-you-go and check the dashboard for any promotional credit attached to a new account.",
      "summary_zh": "官方定价页为每个在服模型列出了每百万 token 的输入与输出价格，既没有描述免费额度，也没有公布请求速率限制。请把 DeepInfra 当作按量付费来看待，新账号是否附带推广额度，到控制台确认。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Listed as metered rather than free because the pricing page we checked documents no free tier; this is not a statement that no signup credit exists.",
      "note_zh": "归到按量计费而不是免费，是因为我们核验的那份定价页没有记录任何免费层；这不等于说注册就一定没有赠送额度。"
    },
    "official_sources": [
      {
        "title": "DeepInfra pricing",
        "url": "https://deepinfra.com/pricing"
      },
      {
        "title": "DeepInfra OpenAI compatibility",
        "url": "https://deepinfra.com/docs/openai_api"
      },
      {
        "title": "DeepInfra models",
        "url": "https://deepinfra.com/models"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "chutes",
    "name": "Chutes",
    "category": "metered-access",
    "signup_url": "https://chutes.ai/app/api",
    "base_url": "https://llm.chutes.ai/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight allows any origin to send an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": false,
    "models": [
      "zai-org/GLM-5",
      "Qwen/Qwen3-32B",
      "unsloth/Mistral-Nemo-Instruct-2407"
    ],
    "limits": {
      "status": "documented-in-plans",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Chutes prices inference per token and sells subscription plans that bundle a daily quota: Plus at $10/month with a bundled daily quota and 6% off pay-as-you-go rates beyond it, and Pro at $20/month with a larger daily quota and 10% off. The pricing page documents no zero-cost tier, and per-plan request-rate numbers are shown on the plan limits page rather than in pricing.",
      "summary_zh": "Chutes 按 token 计费，并出售包含每日额度的订阅计划：Plus 每月 10 美元，含一份每日额度，超出部分在按量价上打 94 折；Pro 每月 20 美元，每日额度更大，超出部分打 9 折。定价页没有记录任何零成本层级，各计划的请求速率数字在计划限额页上，而不在定价页上。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Models are served in TEE (trusted execution environment) variants, which is the differentiator rather than price.",
      "note_zh": "模型以 TEE（可信执行环境）变体提供，这才是它的差异点，而不是价格。"
    },
    "official_sources": [
      {
        "title": "Chutes pricing",
        "url": "https://chutes.ai/pricing"
      },
      {
        "title": "Chutes documentation",
        "url": "https://chutes.ai/docs"
      },
      {
        "title": "Chutes app",
        "url": "https://chutes.ai/app"
      }
    ],
    "source_checked_at": "2026-08-22",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  },
  {
    "id": "scaleway",
    "name": "Scaleway Generative APIs",
    "category": "metered-access",
    "signup_url": "https://console.scaleway.com/",
    "base_url": "https://api.scaleway.ai/v1",
    "openai_compatible": true,
    "browser_check": "supported",
    "browser_check_note": "The CORS preflight echoes the requesting origin and allows an Authorization header, so a browser can call this endpoint directly.",
    "browser_checked_at": "2026-07-25",
    "credit_card_required": true,
    "models": [
      "llama-3.3-70b-instruct",
      "gpt-oss-120b",
      "qwen3-coder-30b-a3b-instruct"
    ],
    "limits": {
      "status": "not_published",
      "requests_per_minute": null,
      "requests_per_day": null,
      "summary": "Every model served through Generative APIs - Serverless is limited by tokens per minute, queries per minute and concurrent requests. Scaleway states that base limits apply only if you have registered a valid payment method, and that they increase automatically if you also verify your identity. The exact numbers live in the Organization quotas page rather than in the rate-limits documentation.",
      "summary_zh": "通过 Generative APIs - Serverless 提供的每个模型，都同时受每分钟 token 数、每分钟查询数和并发请求数三重限制。Scaleway 说明基础限额只在你登记了有效支付方式之后才适用，若再完成身份验证还会自动提高。具体数字在组织配额页面上，而不在限流文档里。"
    },
    "availability": {
      "status": "active",
      "accepting_new_users": true,
      "retires_at": null,
      "note": "Requests sent through the Batches API are not rate limited and are billed at a 50% discount, which is the documented escape hatch for non-real-time work.",
      "note_zh": "走 Batches API 的请求不受限流，并且按五折计费，这是官方给非实时任务留的口子。"
    },
    "official_sources": [
      {
        "title": "Scaleway Generative APIs rate limits",
        "url": "https://www.scaleway.com/en/docs/generative-apis/reference-content/rate-limits/"
      },
      {
        "title": "Scaleway Generative APIs quickstart",
        "url": "https://www.scaleway.com/en/docs/generative-apis/quickstart/"
      },
      {
        "title": "Scaleway Generative APIs pricing",
        "url": "https://www.scaleway.com/en/docs/generative-apis/faq/"
      }
    ],
    "source_checked_at": "2026-07-25",
    "probe": {
      "endpoint_reachable": null,
      "credential_valid": null,
      "sample_quota_exhausted": null,
      "model_available": null,
      "http_status": null,
      "latency_ms": null,
      "checked_at": null,
      "classification": "not-checked",
      "explanation": "No authenticated probe has been published."
    }
  }
]