{
  "source": "AIMap",
  "schema": "free-api/v1",
  "updated": "2026-10-22",
  "generatedAt": "2026-10-22T12:12:50.945Z",
  "note": "Providers that let users call model APIs for free (official channels only). freeType: permanent | new-user-credit | monthly-credit | trial | promo | none. status: active | gone (no longer free). Numbers are copied from official docs only; null = not published.",
  "pages": {
    "en": "https://aimap.page/en/free-models/api/",
    "zh-hans": "https://aimap.page/zh-hans/free-models/api/"
  },
  "providers": [
    {
      "id": "google-gemini",
      "name": {
        "zh-hans": "Google AI Studio（Gemini API）",
        "en": "Google AI Studio (Gemini API)"
      },
      "kind": "vendor",
      "origin": "intl",
      "freeType": "permanent",
      "models": [
        {
          "n": "Gemini 3.8 Flash",
          "t": "text"
        },
        {
          "n": "Gemini 3.6 Flash",
          "t": "text"
        },
        {
          "n": "Gemini 3.5 Flash-Lite",
          "t": "text"
        },
        {
          "n": "Gemini 2.5 Pro",
          "t": "text"
        },
        {
          "n": "Gemini 2.5 Flash",
          "t": "text"
        },
        {
          "n": "Gemma 4",
          "t": "text"
        },
        {
          "n": "Gemini Embedding 2",
          "t": "embedding"
        },
        {
          "n": "Gemini 3.8 Flash TTS",
          "t": "audio"
        }
      ],
      "types": [
        "text",
        "vision",
        "audio",
        "embedding"
      ],
      "quota": {
        "zh-hans": "免费层的输入、输出 token 都不收费；Gemini 4 Argon、3.1 Pro Preview、Nano Banana 图像生成、Veo 等不在免费层。",
        "en": "Free tier: input and output tokens are free of charge. Gemini 4 Argon, 3.1 Pro Preview, Nano Banana image generation, Veo etc. are not on the free tier."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "按项目、按模型限流，具体数值只在 AI Studio 里显示，官方不公开固定数值也不保证；每日额度太平洋时间 0 点重置。",
          "en": "Limited per project and per model; exact values are shown only in AI Studio, with no fixed public numbers or guarantee. Daily quota resets at midnight Pacific."
        }
      },
      "card": false,
      "realName": false,
      "openai": true,
      "apiBase": "https://generativelanguage.googleapis.com/v1beta/openai/",
      "training": {
        "zh-hans": "会：免费层内容会被用于改进 Google 产品。",
        "en": "Yes: free-tier content is used to improve Google products."
      },
      "trainingFlag": "yes",
      "region": {
        "zh-hans": "中国大陆和香港不在官方可用地区列表里。",
        "en": "Mainland China and Hong Kong are not on the official availability list."
      },
      "commercial": {
        "zh-hans": "条款未单独限制，以 Gemini API 附加条款为准。",
        "en": "Not separately restricted; see the Gemini API Additional Terms."
      },
      "docs": [
        {
          "title": "Gemini API 价格（含免费层）",
          "url": "https://ai.google.dev/gemini-api/docs/pricing"
        },
        {
          "title": "Gemini API 速率限制",
          "url": "https://ai.google.dev/gemini-api/docs/rate-limits"
        },
        {
          "title": "可用地区",
          "url": "https://ai.google.dev/gemini-api/docs/available-regions"
        }
      ],
      "pick": {
        "zh-hans": "免费模型档次最高：Gemini 2.5 Pro、3.8 Flash 都能免费调，还带多模态和语音。",
        "en": "Highest-tier free models: Gemini 2.5 Pro and 3.8 Flash, plus multimodal and speech."
      },
      "rank": 1,
      "status": "active",
      "verified": "2026-10-22",
      "note": null
    },
    {
      "id": "zhipu",
      "name": {
        "zh-hans": "智谱 BigModel（GLM Flash 免费模型）",
        "en": "Zhipu BigModel (GLM Flash free models)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "permanent",
      "models": [
        {
          "n": "GLM-4.7-Flash",
          "t": "text"
        },
        {
          "n": "GLM-4-Flash-250414",
          "t": "text"
        },
        {
          "n": "GLM-Z1-Flash",
          "t": "text"
        },
        {
          "n": "GLM-4.6V-Flash",
          "t": "vision"
        },
        {
          "n": "GLM-4.1V-Thinking-Flash",
          "t": "vision"
        },
        {
          "n": "GLM-4V-Flash",
          "t": "vision"
        },
        {
          "n": "CogView-3-Flash",
          "t": "image"
        },
        {
          "n": "CogVideoX-Flash",
          "t": "video"
        }
      ],
      "types": [
        "text",
        "vision",
        "image",
        "video"
      ],
      "quota": {
        "zh-hans": "官方价格表把这 8 个模型的输入、输出都标为“免费”，没有总量上限的说明。",
        "en": "The official price list marks input and output of these 8 models as “free”, with no stated total cap."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "按模型限制并发数，具体数值看控制台“速率限制”页，官方文档未列出免费模型的固定并发。",
          "en": "Per-model concurrency limits shown in the console rate-limit page; the docs list no fixed numbers for free models."
        }
      },
      "openai": true,
      "apiBase": "https://open.bigmodel.cn/api/paas/v4/",
      "docs": [
        {
          "title": "API 定价",
          "url": "https://docs.bigmodel.cn/cn/guide/start/pricing"
        },
        {
          "title": "速率限制",
          "url": "https://docs.bigmodel.cn/cn/api/rate-limit"
        },
        {
          "title": "GLM-4.7-Flash",
          "url": "https://docs.bigmodel.cn/cn/guide/models/free/glm-4.7-flash"
        }
      ],
      "pick": {
        "zh-hans": "国内最大方的永久免费：文本、视觉理解、文生图、文生视频都有免费模型。",
        "en": "Most generous permanent free tier in China: free text, vision, image and video models."
      },
      "rank": 2,
      "status": "active",
      "verified": "2026-10-22",
      "card": null,
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "groq",
      "name": {
        "zh-hans": "Groq（免费计划）",
        "en": "Groq (Free plan)"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "permanent",
      "models": [
        {
          "n": "openai/gpt-oss-120b",
          "t": "text"
        },
        {
          "n": "openai/gpt-oss-20b",
          "t": "text"
        },
        {
          "n": "qwen/qwen3.8-27b",
          "t": "text"
        },
        {
          "n": "whisper-large-v3-turbo",
          "t": "audio"
        },
        {
          "n": "canopylabs/orpheus（TTS）",
          "t": "audio"
        }
      ],
      "types": [
        "text",
        "audio"
      ],
      "quota": {
        "zh-hans": "免费计划可直接调用，按模型限流；升级 Developer 计划才需要绑支付方式。",
        "en": "Usable on the Free plan with per-model limits; a payment method is only needed to upgrade to Developer."
      },
      "limits": {
        "rpm": "30",
        "rpd": "1,000",
        "tpm": "8K",
        "note": {
          "zh-hans": "gpt-oss-120b / 20b、qwen3.8-27b：每分钟 30 次、每天 1,000 次、每分钟 8K token、每天 20 万 token；Whisper 每天 2,000 次。限额按组织计算。",
          "en": "gpt-oss-120b/20b, qwen3.8-27b: 30 RPM, 1,000 RPD, 8K TPM, 200K TPD; Whisper 2,000 RPD. Limits are per organisation."
        }
      },
      "card": false,
      "openai": true,
      "apiBase": "https://api.groq.com/openai/v1",
      "docs": [
        {
          "title": "Rate Limits",
          "url": "https://console.groq.com/docs/rate-limits"
        },
        {
          "title": "Billing FAQs",
          "url": "https://console.groq.com/docs/billing-faqs"
        },
        {
          "title": "Supported Models",
          "url": "https://console.groq.com/docs/models"
        }
      ],
      "pick": {
        "zh-hans": "速度最快，限额写得清清楚楚：每天 1,000 次，不用绑卡。",
        "en": "Fastest, with clearly published limits: 1,000 requests/day, no card."
      },
      "rank": 3,
      "status": "active",
      "verified": "2026-10-22",
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "openrouter",
      "name": {
        "zh-hans": "OpenRouter（:free 模型）",
        "en": "OpenRouter (:free models)"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "permanent",
      "models": [
        {
          "n": "Nemotron 3 Ultra 550B",
          "t": "text"
        },
        {
          "n": "Nemotron 3.5 Super VL",
          "t": "vision"
        },
        {
          "n": "Gemma 4 31B",
          "t": "text"
        },
        {
          "n": "MiniMax M3.1 Flash",
          "t": "text"
        },
        {
          "n": "Inkling",
          "t": "text"
        },
        {
          "n": "Laguna S 2.1",
          "t": "text"
        },
        {
          "n": "North Mini Code",
          "t": "text"
        }
      ],
      "types": [
        "text",
        "vision"
      ],
      "quota": {
        "zh-hans": "模型 ID 以 :free 结尾的模型免费；2026-10-22 模型接口里共 18 个 :free 模型和一个自动挑免费模型的 openrouter/free。",
        "en": "Models whose ID ends in :free cost nothing; on 2026-10-22 the models API lists 18 :free models plus the openrouter/free router."
      },
      "limits": {
        "rpm": "20",
        "rpd": "50（累计充值满 10 美元后 1,000）",
        "tpm": null,
        "note": {
          "zh-hans": "累计充值不足 10 美元：每分钟 20 次、每天 50 次；累计充值满 10 美元：每天 1,000 次。账户余额为负时免费模型也会报错。",
          "en": "Under $10 purchased all-time: 20 RPM, 50 RPD; $10+ purchased: 1,000 RPD. A negative balance blocks free models too."
        }
      },
      "card": false,
      "openai": true,
      "apiBase": "https://openrouter.ai/api/v1",
      "training": {
        "zh-hans": "取决于背后的提供方；账户设置里可对免费模型单独关闭“允许训练的提供方”。",
        "en": "Depends on the upstream provider; account settings let you exclude providers that train, separately for free models."
      },
      "trainingFlag": "depends",
      "docs": [
        {
          "title": "API 限额（免费模型速率）",
          "url": "https://openrouter.ai/docs/api/reference/limits"
        },
        {
          "title": "提供方日志与训练政策",
          "url": "https://openrouter.ai/docs/guides/privacy/provider-logging"
        },
        {
          "title": "模型列表 API",
          "url": "https://openrouter.ai/api/v1/models"
        }
      ],
      "pick": {
        "zh-hans": "一个 Key 用十几个免费模型，规则写得最清楚：每天 50 次，充 10 美元升到 1,000 次。",
        "en": "One key for a dozen-plus free models with the clearest rules: 50/day, 1,000/day after $10 in credits."
      },
      "rank": 4,
      "liveCount": "openrouter",
      "status": "active",
      "verified": "2026-10-22",
      "realName": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "siliconflow",
      "name": {
        "zh-hans": "硅基流动 SiliconFlow（免费模型）",
        "en": "SiliconFlow (free models)"
      },
      "kind": "platform",
      "origin": "cn",
      "freeType": "permanent",
      "models": [
        {
          "n": "XingChenAGI/Xing4.0-29B",
          "t": "text"
        },
        {
          "n": "tencent/Hunyuan-MT-7B",
          "t": "text"
        },
        {
          "n": "PaddlePaddle/PaddleOCR-VL-1.5",
          "t": "vision"
        },
        {
          "n": "BAAI/bge-m3",
          "t": "embedding"
        },
        {
          "n": "BAAI/bge-reranker-v2-m3",
          "t": "rerank"
        },
        {
          "n": "Kwai-Kolors/Kolors",
          "t": "image"
        },
        {
          "n": "XingChenASR V3.2",
          "t": "audio"
        }
      ],
      "types": [
        "text",
        "vision",
        "embedding",
        "rerank",
        "image",
        "audio"
      ],
      "quota": {
        "zh-hans": "价格页标“免费”的模型调用不收费（账单显示 0 元）；付费版同名模型加 Pro/ 前缀。",
        "en": "Models marked “free” on the pricing page cost nothing (billed at ¥0); paid twins carry a Pro/ prefix."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "免费模型的速率限制是固定值，具体数值看模型广场里各模型页。",
          "en": "Free models have fixed rate limits shown on each model page in the console."
        }
      },
      "realName": null,
      "openai": true,
      "apiBase": "https://api.siliconflow.cn/v1",
      "docs": [
        {
          "title": "模型价格",
          "url": "https://www.siliconflow.cn/pricing"
        },
        {
          "title": "Rate Limits",
          "url": "https://docs.siliconflow.com/cn/userguide/rate-limits/rate-limit-and-upgradation"
        }
      ],
      "pick": {
        "zh-hans": "国内一站式免费小模型：文本、OCR、向量、重排、文生图、语音都有免费款。",
        "en": "Free small models for every task in China – text, OCR, embeddings, rerank, image and speech."
      },
      "rank": 5,
      "status": "active",
      "verified": "2026-10-22",
      "card": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "mistral",
      "name": {
        "zh-hans": "Mistral（La Plateforme 免费模式）",
        "en": "Mistral (Free mode)"
      },
      "kind": "vendor",
      "origin": "intl",
      "freeType": "permanent",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "新组织默认处于免费模式（Free mode），可以创建 API Key，在“每月包含用量”内免费调用。",
        "en": "New organisations start in Free mode: create API keys and use the included monthly usage for free."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "每秒请求数、每分钟 token、每月 token 都在后台 Limits 页显示，官方未公开固定数值；可用模型清单也以后台为准。",
          "en": "RPS, tokens/min and tokens/month appear on the Admin Limits page; no fixed public numbers, and the model list is shown in the console."
        }
      },
      "openai": false,
      "apiBase": "https://api.mistral.ai/v1",
      "training": {
        "zh-hans": "会：免费模式下输入和输出可能被用于训练，可随时申请退出。",
        "en": "Yes: in Free mode inputs and outputs may be used for training; you can opt out at any time."
      },
      "trainingFlag": "optout",
      "docs": [
        {
          "title": "Usage and limits",
          "url": "https://docs.mistral.ai/admin/billing-usage/usage-limits"
        },
        {
          "title": "Subscriptions",
          "url": "https://docs.mistral.ai/admin/billing-usage/subscriptions"
        },
        {
          "title": "数据是否用于训练",
          "url": "https://help.mistral.ai/en/articles/347617-do-you-use-my-user-data-to-train-your-artificial-intelligence-models"
        }
      ],
      "note": {
        "zh-hans": "接口格式与 OpenAI 相近但不是官方声明的兼容接口。",
        "en": "API is similar to OpenAI’s but not officially declared compatible."
      },
      "rank": 6,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "commercial": null,
      "region": null
    },
    {
      "id": "cloudflare",
      "name": {
        "zh-hans": "Cloudflare Workers AI",
        "en": "Cloudflare Workers AI"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "permanent",
      "models": [
        {
          "n": "@cf/zai-org/glm-4.7-flash",
          "t": "text"
        },
        {
          "n": "@cf/google/gemma-4-26b-a4b-it",
          "t": "text"
        },
        {
          "n": "@cf/nvidia/nemotron-3-120b-a12b",
          "t": "text"
        },
        {
          "n": "Embedding / 图像 / 语音模型",
          "t": "embedding"
        }
      ],
      "types": [
        "text",
        "embedding",
        "image",
        "audio"
      ],
      "quota": {
        "zh-hans": "Workers 免费计划每天 10,000 Neurons（所有模型共用，UTC 0 点重置）；2026-07-28 起少数高消耗模型只对付费计划开放。",
        "en": "10,000 Neurons per day on Workers Free (shared across models, resets 00:00 UTC). Since 2026-07-28 a few resource-heavy models need Workers Paid."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "额度按 Neurons 计，不同模型折算的 token 数不同，见价格页。",
          "en": "Metered in Neurons; token equivalents differ per model (see pricing)."
        }
      },
      "card": false,
      "openai": true,
      "apiBase": "https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/v1",
      "docs": [
        {
          "title": "Workers AI 价格",
          "url": "https://developers.cloudflare.com/workers-ai/platform/pricing/"
        },
        {
          "title": "部分模型需 Workers Paid（2026-07-28）",
          "url": "https://developers.cloudflare.com/changelog/post/2026-07-28-models-require-workers-paid/"
        }
      ],
      "rank": 7,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "aliyun-bailian",
      "name": {
        "zh-hans": "阿里云百炼（通义千问）",
        "en": "Alibaba Cloud Model Studio (Qwen)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "new-user-credit",
      "models": [
        {
          "n": "Qwen 系列（各模型独立额度）",
          "t": "text"
        },
        {
          "n": "Qwen-VL",
          "t": "vision"
        }
      ],
      "types": [
        "text",
        "vision",
        "image",
        "audio"
      ],
      "quota": {
        "zh-hans": "首次开通百炼自动发放各模型独立的新人免费额度，有效期 90 天；仅华北 2（北京）地域有。",
        "en": "On first activation each model gets its own new-user quota, valid 90 days; Beijing region only."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "只抵扣实时推理，不含 OpenAI 兼容 Batch；未实名用户强制开启“免费额度用完即停”。",
          "en": "Real-time inference only (not OpenAI-compatible Batch); unverified users have “stop when free quota runs out” forced on."
        }
      },
      "realName": {
        "zh-hans": "领取和使用免费额度不用实名；用完后想按量付费需要实名。",
        "en": "Not needed to get or use the free quota; needed for pay-as-you-go afterwards."
      },
      "openai": true,
      "apiBase": "https://dashscope.aliyuncs.com/compatible-mode/v1",
      "docs": [
        {
          "title": "新人免费额度",
          "url": "https://help.aliyun.com/zh/model-studio/new-free-quota"
        }
      ],
      "rank": 8,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "qianfan",
      "name": {
        "zh-hans": "百度千帆（ERNIE）",
        "en": "Baidu Qianfan (ERNIE)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "new-user-credit",
      "models": [
        {
          "n": "ERNIE-4.5-Turbo-128K",
          "t": "text"
        },
        {
          "n": "ERNIE-4.5-Turbo-VL",
          "t": "vision"
        },
        {
          "n": "ERNIE-X1-Turbo-32K",
          "t": "text"
        },
        {
          "n": "DeepSeek-V3.1",
          "t": "text"
        },
        {
          "n": "Qwen3-Coder-480B",
          "t": "text"
        },
        {
          "n": "Kimi-K2-Instruct",
          "t": "text"
        },
        {
          "n": "bge-large-zh",
          "t": "embedding"
        }
      ],
      "types": [
        "text",
        "vision",
        "embedding"
      ],
      "quota": {
        "zh-hans": "首次开通千帆自动发放：17 个指定模型各 100 万 tokens，有效期 3 个月，只能抵扣在线推理，不含批量推理。另有 2025 年 4 月公告的 ERNIE Speed / Lite 免费开放（当前模型列表未再列出，待核实）。",
        "en": "On first activation: 1M tokens for each of 17 listed models, valid 3 months, online inference only. ERNIE Speed / Lite were announced free in Apr 2025 (no longer shown in the current model list – unverified)."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "流控按模型，见模型列表“默认流控”（例如 ERNIE 4.5 Turbo：RPM 5000、TPM 40 万）。",
          "en": "Per-model defaults in the model list (e.g. ERNIE 4.5 Turbo: 5,000 RPM, 400K TPM)."
        }
      },
      "realName": null,
      "openai": true,
      "apiBase": "https://qianfan.baidubce.com/v2",
      "docs": [
        {
          "title": "新用户免费额度",
          "url": "https://cloud.baidu.com/doc/qianfan/s/Imi2rpirg"
        },
        {
          "title": "模型列表（含默认流控）",
          "url": "https://cloud.baidu.com/doc/qianfan/s/rmh4stp0j"
        },
        {
          "title": "部分 ERNIE 模型免费开放公告",
          "url": "https://cloud.baidu.cn/doc/WENXINWORKSHOP/s/wlwg8f1i3"
        }
      ],
      "rank": 9,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "volc-ark",
      "name": {
        "zh-hans": "火山方舟（豆包）",
        "en": "Volcengine Ark (Doubao)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "new-user-credit",
      "models": [
        {
          "n": "豆包系列等方舟在线推理模型",
          "t": "text"
        }
      ],
      "types": [
        "text",
        "vision"
      ],
      "quota": {
        "zh-hans": "注册火山引擎即获赠免费推理额度，各模型分别计算、主账号下共享；具体额度在控制台“开通管理”查看。",
        "en": "Registering a Volcengine account grants free inference quota per model, shared across the main account; amounts are shown in the console."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "只抵扣按 token 后付费的在线推理，不含插件、知识库、批量推理；可开“安心体验模式”，额度用完自动停，不扣费。",
          "en": "Covers pay-per-token online inference only (not plugins, knowledge bases or batch). “Safe trial mode” stops service before any charge."
        }
      },
      "realName": {
        "zh-hans": "领取不用；额度用完后继续用需要实名并开通服务。",
        "en": "Not to claim; needed to continue after the quota runs out."
      },
      "openai": true,
      "apiBase": "https://ark.cn-beijing.volces.com/api/v3",
      "docs": [
        {
          "title": "免费推理额度（更新 2026-08-24）",
          "url": "https://www.volcengine.com/docs/82379/1399514"
        }
      ],
      "rank": 10,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "tencent-tokenhub",
      "name": {
        "zh-hans": "腾讯云 TokenHub（混元等）",
        "en": "Tencent Cloud TokenHub (Hunyuan etc.)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "new-user-credit",
      "models": [
        {
          "n": "Hy3 preview",
          "t": "text"
        },
        {
          "n": "全部语言模型",
          "t": "text"
        },
        {
          "n": "全部多模态理解模型",
          "t": "vision"
        },
        {
          "n": "混元生 3D（积分）",
          "t": "image"
        }
      ],
      "types": [
        "text",
        "vision",
        "image"
      ],
      "quota": {
        "zh-hans": "每个主账号一次：所有语言模型和多模态理解模型各 100 万 tokens，有效期 1 年；活动截至 2026-12-31。原 hunyuan-lite 已于 2026-06-22 下线。",
        "en": "Once per main account: 1M tokens for every language and multimodal-understanding model, valid 1 year; offer runs until 2026-12-31. The old hunyuan-lite was retired on 2026-06-22."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "额度用完后若没开后付费，服务自动停止。",
          "en": "If pay-as-you-go is off, service stops when the quota is used up."
        }
      },
      "openai": true,
      "apiBase": "https://tokenhub.tencentmaas.com/v1",
      "until": "2026-12-31",
      "docs": [
        {
          "title": "TokenHub 新人免费体验包（更新 2026-09-04）",
          "url": "https://cloud.tencent.cn/document/product/1823/130053"
        },
        {
          "title": "混元旧版本模型下线迁移公告",
          "url": "https://cloud.tencent.com/document/product/1729/131925"
        }
      ],
      "rank": 11,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "cohere",
      "name": {
        "zh-hans": "Cohere（试用 Key）",
        "en": "Cohere (trial key)"
      },
      "kind": "vendor",
      "origin": "intl",
      "freeType": "permanent",
      "models": [
        {
          "n": "Command A",
          "t": "text"
        },
        {
          "n": "Command A Reasoning",
          "t": "text"
        },
        {
          "n": "Command A Vision",
          "t": "vision"
        },
        {
          "n": "Command R+",
          "t": "text"
        },
        {
          "n": "North Mini Code",
          "t": "text"
        },
        {
          "n": "Embed",
          "t": "embedding"
        },
        {
          "n": "Rerank",
          "t": "rerank"
        }
      ],
      "types": [
        "text",
        "vision",
        "embedding",
        "rerank"
      ],
      "quota": {
        "zh-hans": "试用（evaluation）Key 免费，每月最多 1,000 次调用。",
        "en": "Trial (evaluation) keys are free, limited to 1,000 API calls per month."
      },
      "limits": {
        "rpm": "20",
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "Chat 每个模型 20 次/分钟；Embed 2,000 条输入/分钟；Rerank 10 次/分钟；每月 1,000 次。",
          "en": "Chat 20 req/min per model; Embed 2,000 inputs/min; Rerank 10 req/min; 1,000 calls/month."
        }
      },
      "openai": true,
      "apiBase": "https://api.cohere.ai/compatibility/v1",
      "commercial": {
        "zh-hans": "试用 Key 面向评估；生产用途需付费 Production Key。",
        "en": "Trial keys are for evaluation; production needs a paid production key."
      },
      "docs": [
        {
          "title": "API Keys and Rate Limits",
          "url": "https://docs.cohere.com/docs/rate-limits"
        },
        {
          "title": "Cohere 价格",
          "url": "https://cohere.com/pricing"
        }
      ],
      "rank": 12,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "training": null,
      "region": null,
      "note": null
    },
    {
      "id": "xfyun",
      "name": {
        "zh-hans": "讯飞星火（Spark Lite）",
        "en": "iFlytek Spark (Lite)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "permanent",
      "models": [
        {
          "n": "Spark Lite",
          "t": "text"
        }
      ],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "官方文档写明 Lite 版“支持免费使用”，需在产品页领取免费额度。",
        "en": "Official docs say Spark Lite “can be used for free”; claim the free quota on the product page."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "最大输入 8K、最大输出 4K；日流控和并发以控制台授权为准。",
          "en": "Max input 8K, max output 4K; daily and concurrency limits per your console authorisation."
        }
      },
      "openai": true,
      "apiBase": "https://spark-api-open.xf-yun.com/v1/",
      "docs": [
        {
          "title": "星火 HTTP 接口文档（兼容 OpenAI SDK）",
          "url": "https://www.xfyun.cn/doc/spark/HTTP%E8%B0%83%E7%94%A8%E6%96%87%E6%A1%A3.html"
        }
      ],
      "rank": 13,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "kimi",
      "name": {
        "zh-hans": "Kimi 开放平台（月之暗面）",
        "en": "Kimi Open Platform (Moonshot AI)"
      },
      "kind": "vendor",
      "origin": "cn",
      "freeType": "new-user-credit",
      "models": [
        {
          "n": "Kimi 系列（不含 Kimi K3）",
          "t": "text"
        }
      ],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "注册赠送 15 元代金券，完成实名认证后发放，3 个月有效；不能用于 Kimi K3。",
        "en": "¥15 voucher on sign-up, issued after real-name verification, valid 3 months; not usable for Kimi K3."
      },
      "realName": true,
      "openai": true,
      "apiBase": "https://api.moonshot.cn/v1",
      "region": {
        "zh-hans": "需中国大陆手机号。",
        "en": "Requires a mainland China phone number."
      },
      "docs": [
        {
          "title": "免费试用额度说明",
          "url": "https://www.kimi.com/help/kimi-api/api-free-trial"
        }
      ],
      "rank": 14,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "note": null
    },
    {
      "id": "nvidia-nim",
      "name": {
        "zh-hans": "NVIDIA NIM（build.nvidia.com）",
        "en": "NVIDIA NIM (build.nvidia.com)"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "permanent",
      "models": [
        {
          "n": "DeepSeek V4 Pro / Flash",
          "t": "text"
        },
        {
          "n": "Kimi K3",
          "t": "text"
        },
        {
          "n": "GLM 5.3 Flash",
          "t": "text"
        },
        {
          "n": "Nemotron 系列",
          "t": "text"
        }
      ],
      "types": [
        "text",
        "vision",
        "embedding"
      ],
      "quota": {
        "zh-hans": "加入免费的 NVIDIA 开发者计划即可调用托管 NIM API，用于原型、研究和测试。",
        "en": "Free hosted NIM API endpoints for NVIDIA Developer Program members, for prototyping, research and testing."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "官方 FAQ：按模型限流、随并发用户数变化，额度在账户页查看；官方文档没有公布固定数值。",
          "en": "Official FAQ: per-model limits that vary with load, shown in your account; no fixed number in the docs."
        }
      },
      "openai": true,
      "apiBase": "https://integrate.api.nvidia.com/v1",
      "commercial": {
        "zh-hans": "只限原型、研究、测试；生产使用需要 NVIDIA AI Enterprise 授权。",
        "en": "Prototyping, research and testing only; production needs an NVIDIA AI Enterprise licence."
      },
      "docs": [
        {
          "title": "Run NIM Anywhere（含价格说明）",
          "url": "https://docs.api.nvidia.com/nim/docs/run-anywhere"
        },
        {
          "title": "NVIDIA NIM FAQ",
          "url": "https://forums.developer.nvidia.com/t/nvidia-nim-faq/300317"
        }
      ],
      "rank": 15,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "training": null,
      "region": null,
      "note": null
    },
    {
      "id": "vercel-ai-gateway",
      "name": {
        "zh-hans": "Vercel AI Gateway",
        "en": "Vercel AI Gateway"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "monthly-credit",
      "models": [
        {
          "n": "Free Tier 可用模型（官方列表）",
          "t": "text"
        }
      ],
      "types": [
        "text",
        "image"
      ],
      "quota": {
        "zh-hans": "每个 Vercel 团队账户每月含 5 美元 AI Gateway 额度，只能用于免费层可用的部分模型。",
        "en": "Every Vercel team gets $5/month of AI Gateway credit, usable on the Free Tier subset of models."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "免费层按模型限流较低，官方未公开具体数值；免费层不能自带 Key（BYOK）。",
          "en": "Lower per-model limits on the free tier (no published numbers); BYOK not available on free tier."
        }
      },
      "openai": true,
      "apiBase": "https://ai-gateway.vercel.sh/v1",
      "docs": [
        {
          "title": "AI Gateway Pricing",
          "url": "https://vercel.com/docs/ai-gateway/pricing"
        }
      ],
      "rank": 16,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "ollama-cloud",
      "name": {
        "zh-hans": "Ollama Cloud",
        "en": "Ollama Cloud"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "monthly-credit",
      "models": [
        {
          "n": "Starter 模型（云端）",
          "t": "text"
        }
      ],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "免费计划含少量起步用量额度，可用 starter 模型；额度数量官方未公开。",
        "en": "Free plan includes starter usage credits for starter models; the amount is not published."
      },
      "openai": true,
      "apiBase": null,
      "note": {
        "zh-hans": "官方 API 地址 https://ollama.com/api；也支持 OpenAI 客户端（功能是子集）。",
        "en": "Native API at https://ollama.com/api; OpenAI clients are also supported (subset)."
      },
      "training": {
        "zh-hans": "不会：官方写明不用云端请求和回复训练模型。",
        "en": "No: Ollama states it does not train on cloud prompts and responses."
      },
      "trainingFlag": "no",
      "docs": [
        {
          "title": "Ollama 价格",
          "url": "https://ollama.com/pricing"
        },
        {
          "title": "Ollama Cloud 文档",
          "url": "https://docs.ollama.com/cloud"
        }
      ],
      "rank": 17,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "limits": null,
      "commercial": null,
      "region": null
    },
    {
      "id": "opencode-zen",
      "name": {
        "zh-hans": "OpenCode Zen（免费模型）",
        "en": "OpenCode Zen (free models)"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "promo",
      "models": [
        {
          "n": "GLM-5.3-Flash Free",
          "t": "text"
        },
        {
          "n": "DeepSeek V4.1 Flash Free",
          "t": "text"
        },
        {
          "n": "MiniMax M3.1 Flash Free",
          "t": "text"
        },
        {
          "n": "MiMo-V2.6-Flash Free",
          "t": "text"
        },
        {
          "n": "Nemotron 3.5 Super VL Free",
          "t": "vision"
        }
      ],
      "types": [
        "text",
        "vision"
      ],
      "quota": {
        "zh-hans": "模型 ID 以 -free 结尾的模型“限时免费”，随时可能下线。",
        "en": "Models whose ID ends in -free are free “for a limited time” and may end any time."
      },
      "openai": true,
      "apiBase": "https://opencode.ai/inference/openai/v1",
      "training": {
        "zh-hans": "因模型而异：GLM-5.3-Flash Free、DeepSeek V4.1 Flash Free 的提供方零留存、不训练；部分模型写明用这段时间收集反馈改进模型。",
        "en": "Varies: GLM-5.3-Flash Free and DeepSeek V4.1 Flash Free providers are zero-retention and do not train; some models state they collect feedback to improve."
      },
      "trainingFlag": "depends",
      "docs": [
        {
          "title": "OpenCode Zen 文档（免费模型说明）",
          "url": "https://opencode.ai/docs/zen/"
        },
        {
          "title": "模型列表 API",
          "url": "https://opencode.ai/zen/v1/models"
        }
      ],
      "rank": 18,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "limits": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "cerebras",
      "name": {
        "zh-hans": "Cerebras（免费试用）",
        "en": "Cerebras (free trial)"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "trial",
      "models": [
        {
          "n": "gpt-oss-120b",
          "t": "text"
        },
        {
          "n": "qwen-3.8-27b",
          "t": "vision"
        }
      ],
      "types": [
        "text",
        "vision"
      ],
      "quota": {
        "zh-hans": "绑定已验证的支付方式后一次性送 5 美元，30 天内有效；官方明确没有长期免费层。",
        "en": "One-time $5 credit after adding a verified payment method, valid 30 days; officially no permanent free tier."
      },
      "limits": {
        "rpm": "5",
        "rpd": null,
        "tpm": "30K",
        "note": {
          "zh-hans": "试用期每分钟 5 次、未缓存 30K TPM、每天 100 万 token。",
          "en": "Trial: 5 RPM, 30K uncached TPM, 1M tokens/day."
        }
      },
      "card": true,
      "openai": true,
      "apiBase": "https://api.cerebras.ai/v1",
      "docs": [
        {
          "title": "Rate Limits（含 Free Trial FAQ）",
          "url": "https://inference-docs.cerebras.ai/support/rate-limits"
        }
      ],
      "rank": 19,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "realName": null,
      "training": null,
      "commercial": null,
      "region": null,
      "note": null
    },
    {
      "id": "modelscope",
      "name": {
        "zh-hans": "魔搭 ModelScope API-Inference",
        "en": "ModelScope API-Inference"
      },
      "kind": "platform",
      "origin": "cn",
      "freeType": "permanent",
      "models": [
        {
          "n": "魔搭上热门开源模型（Qwen、DeepSeek、GLM 等，标有 API-Inference 图标）",
          "t": "text"
        }
      ],
      "types": [
        "text",
        "vision",
        "image"
      ],
      "quota": {
        "zh-hans": "调用用“魔粒”兑换：每天登录送 200 魔粒，绑定阿里云账号再加 50（当日有效）；轻量模型 0.5 魔粒/次、主流 1、旗舰 2。按此折算每天约 125–500 次。",
        "en": "Calls are paid in Magicubes: 200/day for logging in, +50/day after linking Alibaba Cloud (valid same day); 0.5 / 1 / 2 per call for light / standard / flagship models – roughly 125–500 calls a day."
      },
      "limits": {
        "rpm": null,
        "rpd": null,
        "tpm": null,
        "note": {
          "zh-hans": "并发按平台负载动态限流，原则是保证单并发可用。",
          "en": "Concurrency is limited dynamically by load, guaranteeing single-request use."
        }
      },
      "realName": true,
      "openai": true,
      "apiBase": "https://api-inference.modelscope.cn/v1",
      "commercial": {
        "zh-hans": "官方定位为非商业、非盈利产品，请勿用于需要高并发或 SLA 的线上任务。",
        "en": "Officially non-commercial and non-profit; do not use for high-concurrency or SLA-bound production."
      },
      "docs": [
        {
          "title": "API-Inference 使用限制",
          "url": "https://modelscope.cn/docs/model-service/API-Inference/limits"
        },
        {
          "title": "魔粒说明",
          "url": "https://modelscope.cn/docs/magicube/intro"
        }
      ],
      "note": {
        "zh-hans": "需要先绑定已实名的阿里云账号。",
        "en": "Requires a linked, ID-verified Alibaba Cloud account."
      },
      "rank": 20,
      "status": "active",
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "training": null,
      "region": null
    },
    {
      "id": "github-models",
      "name": {
        "zh-hans": "GitHub Models",
        "en": "GitHub Models"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "none",
      "status": "gone",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "2026-07-30 起已全面退役，Playground、模型目录和推理 API 都停了。",
        "en": "Fully retired on 2026-07-30: playground, catalog and inference API are gone."
      },
      "docs": [
        {
          "title": "GitHub Models 已退役",
          "url": "https://docs.github.com/en/github-models/use-github-models/prototyping-with-ai-models"
        }
      ],
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "openai": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "apiBase": null,
      "region": null,
      "note": null
    },
    {
      "id": "hf-inference",
      "name": {
        "zh-hans": "Hugging Face Inference Providers",
        "en": "Hugging Face Inference Providers"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "none",
      "status": "gone",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "免费用户不再有每月额度，需要购买额度或升级 PRO。",
        "en": "Free users get no monthly credits; buy credits or upgrade to PRO."
      },
      "docs": [
        {
          "title": "Pricing and Billing",
          "url": "https://huggingface.co/docs/inference-providers/pricing"
        }
      ],
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "openai": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "apiBase": null,
      "region": null,
      "note": null
    },
    {
      "id": "sambanova",
      "name": {
        "zh-hans": "SambaNova Cloud",
        "en": "SambaNova Cloud"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "none",
      "status": "gone",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "免费计划需要先绑支付方式并购买额度才能发出第一次请求。",
        "en": "The Free plan requires adding a payment method and buying credits before your first request."
      },
      "docs": [
        {
          "title": "SambaNova Plans",
          "url": "https://cloud.sambanova.ai/plans"
        }
      ],
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "openai": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "apiBase": null,
      "region": null,
      "note": null
    },
    {
      "id": "together",
      "name": {
        "zh-hans": "Together AI",
        "en": "Together AI"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "none",
      "status": "gone",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "官方写明目前没有免费试用，首次至少充值 15 美元。",
        "en": "No free trial; first purchase must be at least $15."
      },
      "docs": [
        {
          "title": "Together 额度说明",
          "url": "https://docs.together.ai/docs/billing"
        }
      ],
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "openai": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "apiBase": null,
      "region": null,
      "note": null
    },
    {
      "id": "fireworks",
      "name": {
        "zh-hans": "Fireworks AI",
        "en": "Fireworks AI"
      },
      "kind": "platform",
      "origin": "intl",
      "freeType": "none",
      "status": "gone",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "预付费积分制，官方文档未写明赠送额度；无支付方式或无余额时全账号限 10 RPM，有支付方式且有余额时上限 6,000 RPM（2026-10-22 核对）。",
        "en": "Prepaid credits; the docs mention no free credit; account-wide 10 RPM without a payment method or credits, 6,000 RPM max with a payment method and credits (checked 2026-10-22)."
      },
      "docs": [
        {
          "title": "Account quotas",
          "url": "https://docs.fireworks.ai/guides/quotas_usage/rate-limits"
        }
      ],
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "openai": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "apiBase": null,
      "region": null,
      "note": null
    },
    {
      "id": "hunyuan-lite",
      "name": {
        "zh-hans": "腾讯混元 hunyuan-lite",
        "en": "Tencent hunyuan-lite"
      },
      "kind": "platform",
      "origin": "cn",
      "freeType": "none",
      "status": "gone",
      "models": [],
      "types": [
        "text"
      ],
      "quota": {
        "zh-hans": "2026-06-22 下线，原混元平台 9 月 30 日停服，改用 TokenHub 新人额度。",
        "en": "Retired on 2026-06-22; the old Hunyuan platform closed on 30 Sep – use TokenHub’s new-user quota."
      },
      "docs": [
        {
          "title": "旧版本模型下线迁移公告",
          "url": "https://cloud.tencent.com/document/product/1729/131925"
        }
      ],
      "verified": "2026-10-22",
      "pick": null,
      "card": null,
      "realName": null,
      "openai": null,
      "limits": null,
      "training": null,
      "commercial": null,
      "apiBase": null,
      "region": null,
      "note": null
    }
  ]
}