| SenseNova · Token Plan 公测免费SenseNova · Token PlanSenseNova 6.8 Flash Lite · SenseNova U1 FastSenseNova 6.8 Flash Lite · SenseNova U1 Fast |
公测期完全免费：Free 档 ¥0/月，60,000 积分 / 5 小时；特殊模型除外；最多 20 个 API Key。Completely free during public beta: 60,000 credits per 5 hours on the Free tier; no top-up or card required |
SenseNova account + API Key · public-beta quotaSenseNova account + API Key · public-beta quota |
Official source ↗
FreeLLM record ↗ |
| OpenCode Zen · free model poolOpenCode Zen · free model poolMiMo-V2.6-Flash Free · Big Pickle · Step 5 Preview Free · Ling 3.1 Flash Free · Ling 3.0 Flash Fin Free · LongCat 2.5 Preview Free · Exo Free · Nemotron · Muse SparkMiMo-V2.6-Flash Free · Big Pickle · Step 5 Preview Free · Ling 3.1 Flash Free · Ling 3.0 Flash Fin Free · LongCat 2.5 Preview Free · Exo Free · Nemotron · Muse Spark |
MiMo-V2.6-Flash Free and other selected Zen models · input, output and cache listed as FreeMiMo-V2.6-Flash Free and other selected Zen models · input, output and cache listed as Free |
Global / OpenCode accountGlobal / OpenCode account |
Official source ↗
FreeLLM record ↗ |
| 阿里云百炼中国站 · 北京地域新人免费额度Offer detailsQwen and other eligible Model Studio real-time inference modelsQwen and other eligible Model Studio real-time inference models |
中国站：仅华北2（北京）符合条件的模型享新人免费额度；通常每模型 1,000,000 Tokens、90 天有效，具体以模型价格页为准。Eligible models usually receive 1,000,000 shared input+output tokens each; exact quota varies by model. OAuth authentication has a separate 2,000 calls/day free allowance. |
中国站百炼 · 北京地域 · 通用 API KeyAPI Key |
Official source ↗
FreeLLM record ↗ |
| Google AI Studio · Gemini API free tierGoogle AI Studio · Gemini API free tierGemini 3.8 Flash (gemini-3.8-flash) and other Gemini models on the free tierGemini 3.8 Flash (gemini-3.8-flash) and other Gemini models on the free tier |
Free tier: input, output and context caching listed as Free of chargeFree tier: input, output and context caching listed as Free of charge |
Google account; availability may vary by regionGoogle account; availability may vary by region |
Official source ↗
FreeLLM record ↗ |
| Groq · free plan rate limitsGroq · free plan rate limitsgpt-oss-120b / gpt-oss-20b · qwen3.8-27b · llama-prompt-guard · whisper-large-v3gpt-oss-120b / gpt-oss-20b · qwen3.8-27b · llama-prompt-guard · whisper-large-v3 |
Free plan with per-model RPM / RPD / TPM limits (e.g. gpt-oss-120b: 30 RPM · 1K RPD · 8K TPM)Free plan with per-model RPM / RPD / TPM limits (e.g. gpt-oss-120b: 30 RPM · 1K RPD · 8K TPM) |
Groq account; rate limits apply at organization levelGroq account; rate limits apply at organization level |
Official source ↗
FreeLLM record ↗ |
| Cerebras · free inference creditsCerebras · free inference creditsAll Cerebras-powered hosted models while free credits lastAll Cerebras-powered hosted models while free credits last |
需绑卡：添加有效付款方式后发放一次性 $5 促销额度，激活后 30 天过期；用完或过期后 API 与 Playground 暂停（不会自动扣费）One-time $5 promotional credit granted after adding a valid payment method |
Cerebras 账号 + 有效付款方式（不会自动扣费）Cerebras account plus a valid payment method on file (no automatic charge) |
Official source ↗
FreeLLM record ↗ |
| OpenRouter · free model variantsOpenRouter · free model variantsModel variants whose ID ends in :free (free model variants)Model variants whose ID ends in :free (free model variants) |
:free variants: 20 req/min; 50 req/day (under 10 credits purchased) or 1000 req/day (10+ credits)free variants: 20 req/min; 50 req/day (under 10 credits purchased) or 1000 req/day (10+ credits) |
OpenRouter account; daily caps depend on lifetime credit purchasesOpenRouter account; daily caps depend on lifetime credit purchases |
Official source ↗
FreeLLM record ↗ |
| Mistral · La Plateforme free modeMistral · La Plateforme free modeMistral API models under free mode rate limitsMistral API models under free mode rate limits |
Free mode (default): lowest rate limits, intended for evaluation and prototypingFree mode (default): lowest rate limits, intended for evaluation and prototyping |
Mistral accountMistral account |
Official source ↗
FreeLLM record ↗ |
| Cohere · free trial API keyCohere · free trial API keyCommand chat models (trial key rate limits)Command chat models (trial key rate limits) |
Trial API key: free, rate-limited, not permitted for production or commercial useTrial API key: free, rate-limited, not permitted for production or commercial use |
Cohere personal accountCohere personal account |
Official source ↗
FreeLLM record ↗ |
| Hugging Face · Inference Providers free tierHugging Face · Inference Providers free tierOpen models served via multi-provider inference (chat, embeddings, image and more)Open models served via multi-provider inference (chat, embeddings, image and more) |
Generous free tier included; additional credits for PRO users and organizationsGenerous free tier included; additional credits for PRO users and organizations |
Hugging Face accountHugging Face account |
Official source ↗
FreeLLM record ↗ |
| SiliconFlow · ¥0 listed modelsSiliconFlow · ¥0 listed modelsXing4.0-29B · PaddleOCR-VL-1.5 · Hunyuan-MT-7B · BAAI embeddings/reranker · Kwai-Kolors (¥0 listed)Xing4.0-29B · PaddleOCR-VL-1.5 · Hunyuan-MT-7B · BAAI embeddings/reranker · Kwai-Kolors (¥0 listed) |
Selected official ¥0 models now include China Telecom Xing4.0-29B alongside OCR, embedding, reranker and image models.Selected official ¥0 models now include China Telecom Xing4.0-29B alongside OCR, embedding, reranker and image models. |
SiliconFlow account; China registrationSiliconFlow account; China registration |
Official source ↗
FreeLLM record ↗ |
| ModelScope · API-Inference free callsModelScope · API-Inference free callsHosted open models (LLM · MLLM · text-to-image) marked with the API-Inference badgeHosted open models (LLM · MLLM · text-to-image) marked with the API-Inference badge |
API-Inference is free for registered users; dynamic rate limits and 魔粒 (MoLi) deduction applyAPI-Inference is free for registered users; dynamic rate limits and MoLi deduction apply |
ModelScope account + Aliyun binding + real-name verificationModelScope account + Aliyun binding + real-name verification |
Official source ↗
FreeLLM record ↗ |
| Agnes AI · 免费/默认多模态 APIAgnes AI · Free/default multimodal APIagnes-3.0-flash · agnes-2.5-flash · agnes-image-2.5-flash · agnes-image-2.1-flash · agnes-image-2.0-flash · agnes-video-2.5-flashagnes-3.0-flash · agnes-2.5-flash · agnes-image-2.5-flash · agnes-image-2.1-flash · agnes-image-2.0-flash · agnes-video-2.5-flash |
Free/default API access for selected Agnes models with public rate limitsFree/default API access for selected Agnes models with public rate limits |
Agnes AI 账号 + API Key；国内站与国际站任一注册即可，两站 Key 通用Agnes AI account and API key; register on either the CN or the global console - the same key works on both |
Official source ↗
FreeLLM record ↗ |
| StepFun · 限时免费 API 模型StepFun · Limited-time free API modelsstepaudio-3-realtime-preview · stepaudio-3-chat-preview · stepaudio-3-gen-preview · stepaudio-3-music-previewstepaudio-3-realtime-preview · stepaudio-3-chat-preview · stepaudio-3-gen-preview · stepaudio-3-music-preview |
StepFun 官方定价页当前将 stepaudio-3-realtime-preview、stepaudio-3-chat-preview、stepaudio-3-gen-preview、stepaudio-3-music-preview 标注为限免中StepFun's official pricing page currently marks stepaudio-3-realtime-preview, stepaudio-3-chat-preview, stepaudio-3-gen-preview and stepaudio-3-music-preview as limited-time free |
StepFun Open Platform account and API keyStepFun Open Platform account and API key |
Official source ↗
FreeLLM record ↗ |
| 小红书 Dots · Dots3-Note Preview APIDots · Dots3-Note Preview APIDots3-Note PreviewDots3-Note Preview |
Dots3-Note Preview 当前免费试用，按官方限速使用60 requests/minute; 1,500,000 tokens/minute |
小红书账号或手机号登录 Dots 国内平台Dots |
Official source ↗
FreeLLM record ↗ |
| 上海人工智能实验室 · Atria Dawn Preview（注册 1 亿 Token）Atria Dawn Preview（ 1 Token）Atria Dawn PreviewAtria Dawn Preview |
注册即领 1 亿 Token 免费额度；控制台当前档位 Atria-Dawn-Preview 输入/输出单价为 $01 Token ； Atria-Dawn-Preview / $0 |
国际控制台需 Google 账号；国内可走 discovery.intern-ai.org.cn Token Plan 入口Google ； discovery.intern-ai.org.cn Token Plan |
Official source ↗
FreeLLM record ↗ |
| AMD Radeon Cloud · Token Factory 免费模型 APIAMD Radeon Cloud · Token Factory APIDeepSeek V4 / V4.1 · Qwen3.8 · MiniCPM5DeepSeek V4 / V4.1 · Qwen3.8 · MiniCPM5 |
6 个当前共享聊天模型 · 无需创建实例 · 不消耗 credits · 最高 1,048,576 tokens 上下文6 current shared chat models · no instance · no credits · up to 1,048,576-token context |
Radeon Cloud account verification + rc- API key; shared rate limits applyRadeon Cloud account verification + rc- API key; shared rate limits apply |
Official source ↗
FreeLLM record ↗ |
| 阶跃星辰 Step Plan · 免费体验已结束StepFun · Step Plan （ ）Step 5 Preview · Step 3.7 Flash · StepAudio 3 familyStep 5 Preview · Step 3.7 Flash · StepAudio 3 family |
已结束：Step Plan 官网已挂出“关于免费体验及 Plan 购买名额调整的通知”，新增免费体验领取通道已关闭；“最高 75 天”攻略不再适用。Ended: the official Step Plan page posted a notice adjusting the free experience and plan quotas; new free-experience claims are closed (no later than 2026-09-22) and the "up to 75 days" guide no longer applies. Step Plan is now a paid subscription. |
StepFun account + Step API Key; Step Plan entitlement requiredStepFun account + Step API Key; Step Plan entitlement required |
Official source ↗
FreeLLM record ↗ |
| Puter.js · MiMo-V2.6-Flash 免费 PlaygroundPuter.js · MiMo-V2.6-Flash PlaygroundXiaomi MiMo-V2.6-FlashXiaomi MiMo-V2.6-Flash |
Puter 账号自带免费 AI allowance；MiMo-V2.6-Flash 可在官方浏览器 Playground 免费试用Every Puter account includes a free AI allowance; MiMo-V2.6-Flash is available in the official browser playground |
Global · Puter account · browser no Xiaomi keyGlobal · Puter account · browser no Xiaomi key |
Official source ↗
FreeLLM record ↗ |
| Tencent Cloud TokenHub · 1M free Tokens per modelTencent Cloud TokenHub · 1M free Tokens per modelAll language and multimodal-understanding modelsAll language and multimodal-understanding models |
Each language and multimodal-understanding model offers 1,000,000 free Tokens for new users; each model's package is claimable once.Each language and multimodal-understanding model offers 1,000,000 free Tokens for new users; each model's package is claimable once. |
Tencent Cloud account + identity verificationTencent Cloud account + identity verification |
Official source ↗
FreeLLM record ↗ |
| 百度千帆 · 新客最高 100 万+ Tokens 免费调用100 + TokensERNIE family + supported preset modelsERNIE family + supported preset models |
百度千帆当前产品页明确标注：新客开通可免费领取最高 1,000,000+ Tokens 调用量；具体模型与额度以控制台为准。1,000,000+ Tokens ； 。 |
百度智能云账号 + 千帆开通 + IAM API Key+ + IAM API Key |
Official source ↗
FreeLLM record ↗ |
| 快手万擎 StreamLake · 当前 2 款 KAT API ¥0StreamLake · 2 KAT API ¥0KAT-Coder-Air-V1 · KAT-Coder-Exp-72B-1010KAT-Coder-Air-V1 · KAT-Coder-Exp-72B-1010 |
StreamLake 万擎官方价格表当前将 KAT-Coder-Air-V1 与 KAT-Coder-Exp-72B-1010 的输入、输出和缓存全部标为“免费”。StreamLake KAT-Coder-Air-V1 KAT-Coder-Exp-72B-1010 、 “ ”。 |
账号 + 实名 + 万擎开通 + API Key / 推理点+ + + API Key |
Official source ↗
FreeLLM record ↗ |
| 天翼云星辰 TokenHub · 支持模型 50 万 Tokens / 14 天TokenHub · 50 Tokens / 14Supported preset modelsSupported preset models |
平台升级后的官方规则：支持免费体验的模型赠送 500,000 Tokens 免费推理额度，自领取起 14 天有效。500,000 Tokens for each supported free-experience model under current v2.0 policy; exact eligible models are dynamic |
星辰 TokenHub + APP Key；建议开启免费额度耗尽自动关闭TokenHub + APP Key； |
Official source ↗
FreeLLM record ↗ |
| 智谱 BigModel · 新用户 2000 万 Tokens 体验包BigModel · 2000 TokensGLM API models covered by the new-user resource packGLM API models covered by the new-user resource pack |
智谱 BigModel 官方注册页当前标注：新用户免费赠送专享 20,000,000 Tokens 体验包。20,000,000 Tokens new-user trial pack; exact model split and expiry are account/resource-pack specific |
BigModel 注册 + API Key；资源包范围需登录确认BigModel + API Key； |
Official source ↗
FreeLLM record ↗ |
| Xiaomi MiMo · V2.5 TTS 三款 API 当前限时免费Xiaomi MiMo · V2.5 TTS APImimo-v2.5-tts · mimo-v2.5-tts-voicedesign · mimo-v2.5-tts-voiceclonemimo-v2.5-tts · mimo-v2.5-tts-voicedesign · mimo-v2.5-tts-voiceclone |
MiMo 官方定价与模型页当前将 V2.5-TTS、VoiceDesign、VoiceClone 三款 API 标为限时免费。Three MiMo V2.5 TTS APIs are temporarily priced at zero on the official Open Platform |
MiMo account + API Key · China / overseas pricing pages both mark TTS freeMiMo account + API Key · China / overseas pricing pages both mark TTS free |
Official source ↗
FreeLLM record ↗ |
| NVIDIA NIM · 当前 38 个 Free EndpointNVIDIA NIM · 38 Free Endpoint38 current Free Endpoint models38 current Free Endpoint models |
NVIDIA API Catalog 当前显示 38 个 Free Endpoint；免费 NVIDIA Developer Program 可访问托管 NIM API endpoints。Free NVIDIA Developer Program access to catalog models currently labeled Free Endpoint |
Free NVIDIA Developer Program + API Key · no card requiredFree NVIDIA Developer Program + API Key · no card required |
Official source ↗
FreeLLM record ↗ |
| 阿里云百炼 decision-model-preview · 北京/新加坡限时免费decision-model-previewdecision-model-previewdecision-model-preview |
百炼官方模型页当前将 decision-model-preview 在北京和新加坡的调用价格都标为“限时免费”。Current zero-price limited-time preview for decision-model-preview in Beijing and Singapore |
Model Studio API Key + Workspace ID · Beijing / SingaporeModel Studio API Key + Workspace ID · Beijing / Singapore |
Official source ↗
FreeLLM record ↗ |
| 火山方舟 · 新用户 50 万 Tokens + 安心体验模式50 Tokens +Supported Ark online inference modelsSupported Ark online inference models |
火山方舟官方安心体验模式：新用户可使用平台赠送的 500,000 Tokens 免费推理额度；额度接近/耗尽后服务自动暂停，避免继续产生模型推理费用。New-user platform gift of 500,000 inference tokens with optional Safe Experience Mode that pauses model service at the free-quota boundary |
Ark account + API Key · Beijing v3 OpenAI-compatible APIArk account + API Key · Beijing v3 OpenAI-compatible API |
Official source ↗
FreeLLM record ↗ |
| Alibaba Cloud Model Studio 国际站 · 新加坡新人免费额度Alibaba Cloud Model StudioEligible Model Studio Singapore-region modelsEligible Model Studio Singapore-region models |
国际站：首次开通 Model Studio 新加坡地域后，平台自动发放各模型新人免费额度；仅新加坡地域享该免费额度，其他国际地域无此新人额度。Singapore-region new-user per-model free quota on Alibaba Cloud International |
Alibaba Cloud International account + required account information + Singapore API KeyAlibaba Cloud International account + required account information + Singapore API Key |
Official source ↗
FreeLLM record ↗ |
| 讯飞星火 Spark Lite · 当前免费 APISpark Lite · APISpark Lite (model=lite)Spark Lite (model=lite) |
讯飞星火官方 HTTP 文档当前明确：Spark Lite 为轻量级大语言模型，支持免费使用。Current free use of the Spark Lite model on iFlytek's official Spark API |
讯飞开放平台 + Spark Lite 授权 + APIPasswordEnable Spark Lite in iFlytek Open Platform and authenticate with the HTTP APIPassword |
Official source ↗
FreeLLM record ↗ |
| 百度 AI Studio 星河 · 每位开发者 100 万免费 TokensAI Studio · 100 TokensERNIE 5.1 / ERNIE 5.0 / ERNIE 4.5 / DeepSeek / Kimi / Qwen modelsERNIE 5.1 / ERNIE 5.0 / ERNIE 4.5 / DeepSeek / Kimi / Qwen models |
百度 AI Studio 星河大模型 API 官方文档：每个开发者提供 1,000,000 免费 Tokens，共享用于不同模型调用。One shared 1,000,000-token developer allowance across AI Studio LLM API models |
AI Studio account + Access TokenAI Studio account + Access Token |
Official source ↗
FreeLLM record ↗ |
| 华为云 ModelArts Studio MaaS · 动态免费模型 APIModelArts Studio MaaS · APIDynamic preset models under Free Service tabDynamic preset models under Free Service tab |
华为云 ModelArts Studio（MaaS）提供动态“免费服务”预置模型，可直接通过 API 推理；免费模型、额度和失效时间会调整。Dynamic preset-model free services with either time-limited unlimited usage or claimed limited quota |
Huawei Cloud + MaaS authorization + region API KeyHuawei Cloud + MaaS authorization + region API Key |
Official source ↗
FreeLLM record ↗ |
| 天翼AI TeleChat3-438B · 1 亿 Token 免费试用 / 3 个月AI TeleChat3-438B · 1 Token / 3TeleChat3-438B-A24BTeleChat3-438B-A24B |
天翼AI官方定价：TeleChat3-438B-A24B 免费试用为 0 元 / 1 亿 Token 包，领取后 3 个月有效。One-time 100,000,000-token free trial pack for TeleChat3-438B-A24B |
TeleAI account + long-lived API KeyTeleAI account + long-lived API Key |
Official source ↗
FreeLLM record ↗ |
| 国家超算互联网 SCNet · 文档解析 API 当前免费SCNet · APIDocument Parsing APIDocument Parsing API |
SCNet 官方活动页明确：文档解析 API 调用限时免费、不限次数；当前文档智能服务页仍将文档解析标记为免费。Current limited-time free promotion for the SCNet Document Parsing API |
SCNet account + universal API KeySCNet account + universal API Key |
Official source ↗
FreeLLM record ↗ |
| 中国联通元景 MaaS · 试用账号 5 次/分钟免费体验MaaS · 5Yuanjing + third-party models available to trial accountsYuanjing + third-party models available to trial accounts |
元景 MaaS 官方多个模型页明确：试用账号每分钟可调用 5 次进行模型体验；付费账号可放开限流。Trial-account model access limited to 5 calls per minute |
Yuanjing MaaS account + access tokenYuanjing MaaS account + access token |
Official source ↗
FreeLLM record ↗ |
| 百川 Baichuan-M3 Plus · 医疗机构定向永久免费 APIBaichuan-M3 Plus · APIBaichuan-M3 PlusBaichuan-M3 Plus |
百川“海纳百川计划”对符合准入条件的医疗服务机构提供 Baichuan-M3 Plus API 永久免费调用。Application-based permanent-free access under the Haina Baichuan medical public-interest program |
仅限指定机构/场景；普通开放平台调用仍按官方价格计费Apply to the Haina Baichuan program, pass review, obtain API access, and comply with medical-use restrictions |
Official source ↗
FreeLLM record ↗ |
| PPIO 派欧云 · 受邀新用户 ¥5 免费 API 额度PPIO · ¥5 APIPPIO Model API catalogPPIO Model API catalog |
PPIO 官方注册/登录页当前显示：受邀新用户注册可获 ¥5 免费额度，页面估算约可调用 500 万 DeepSeek-V3 Tokens。Invite-gated one-time account credit for new PPIO users |
需通过官方邀请注册入口/邀请码完成注册Register through an official PPIO invite link/code, create an API key, then call the OpenAI-compatible API |
Official source ↗
FreeLLM record ↗ |
| 无问芯穹 GenStudio · Embedding / Rerank API 当前免费GenStudio · Embedding / Rerank APIbge-m3 · bge-reranker-v2-m3bge-m3 · bge-reranker-v2-m3 |
无问芯穹官方 FAQ 当前明确：GenStudio 向量嵌入模型 API 与重排序模型 API 仍暂不收费；基础版 LLM API 已于 2026-03-30 停止免费。Current no-charge policy for GenStudio retrieval APIs (Embeddings and Rerank) |
GenStudio API Key · cloud.infini-ai.comGenStudio API Key · cloud.infini-ai.com |
Official source ↗
FreeLLM record ↗ |
| 蚂蚁百灵 Ling-3.1-flash · 限时免费 APILing-3.1-flash · APILing-3.1-flash · 560B 总参数 / 25B 激活参数Ant Ling Ling-3.1-flash · Vercel AI Gateway |
Vercel AI Gateway 的 inclusionai/ling-3.1-flash-free 路由限时免费至 2026-10-13The inclusionai/ling-3.1-flash-free route on Vercel AI Gateway is free through 2026-10-13 |
需要 Vercel AI Gateway 账户与 API Key；此处免费规则仅适用于 Vercel 路由Vercel AI Gateway account and API key required; the free terms apply only to this route |
Official source ↗
FreeLLM record ↗ |
| Laya 结构化判定模型 · 限时免费 APILaya · APILaya · System One evaluation modelLaya · System One evaluation model |
Vercel AI Gateway 的 convaiinnovations/laya-free 当前免费至 2026-10-31；用于结构化判定，不是通用聊天模型convaiinnovations/laya-free is free on Vercel AI Gateway through 2026-10-31; it performs structured evaluation rather than general chat |
需要 Vercel AI Gateway API Key；通过 AI SDK experimental_evaluate 调用Vercel AI Gateway API key required; call through AI SDK experimental_evaluate |
Official source ↗
FreeLLM record ↗ |
| 云知声 U2 Flash · 国庆免费月 APIU2 Flash · APIU2 Flash (u2-flash)U2 Flash (u2-flash) |
云知声官方价格页：U2 Flash 活动时间 2026-09-30 至 2026-10-31，输入、缓存命中、输出价格均为 0 元/百万 Tokens；官网横幅写明“限时免费无限量调用”。Limited-time ¥0 pricing for U2 Flash input, cached input and output on the official Unisound MaaS API |
云知声 MaaS 账号 + API KeyRegister on Unisound MaaS, create an API key, call model u2-flash |
Official source ↗
FreeLLM record ↗ |
| 书生·端砚 墨点计划 · 测试期每月免费 TokenToken书生 Intern-S2 / Atria-Dawn-Preview / Agents-A1 + 第三方模型（按墨点计费）Intern InkStone Token Plan monthly free quota |
上海 AI 实验室官方公告：书生·端砚“墨点计划”（Token Plan）测试期间，每位用户每月可获免费 Token，用于调用多元模型与科研服务。AI ： · “ ”（Token Plan） ， Token， 。 |
书生·端砚账号 + Token Plan API Key+ Token Plan API Key |
Official source ↗
FreeLLM record ↗ |
| 阶跃星辰 Step 5 Preview · 一周限免（OpenCode Zen 等）Step 5 Preview · （OpenCode Zen ）Step 5 Preview · 600B 总参数 / 27B 激活 · 1M 上下文StepFun Step 5 Preview · one-week free access (OpenCode Zen and partner tools) |
阶跃星辰旗舰 Step 5 Preview 自 2026-10-08 起在 OpenCode Zen（step-5-preview-free）等合作工具限时免费约一周StepFun's flagship Step 5 Preview is free for about one week from 2026-10-08 via OpenCode Zen (step-5-preview-free) and partner tools |
需要 OpenCode Zen 账号与 API Key（或在 Cline / Kilo Code / Nous Portal 等合作工具内选择免费模型）OpenCode Zen account and API key, or select the free model inside partner tools such as Cline, Kilo Code or Nous Portal |
Official source ↗
FreeLLM record ↗ |
| GLM-4.7-Flash · BigModel 官方免费 APIGLM-4.7-Flash · BigModel APIGLM-4.7-FlashGLM-4.7-Flash |
智谱官方发布公告明确 GLM-4.7-Flash 在 BigModel.cn 上线并供免费调用。Official BigModel free-call route for GLM-4.7-Flash; provider may change limits or policy. |
BigModel account + API KeyBigModel account + API Key |
Official source ↗
FreeLLM record ↗ |
| Xiaomi MiMo · 新用户 ¥10 API 体验金Xiaomi MiMo · ¥10 APIMiMo API modelsMiMo API models |
符合当前官方邀请规则的新用户完成绑定后双方各得 ¥10 赠金；仅抵扣 API 按量调用。One-time referral/new-user API credit, not a permanently free model. |
China-registered MiMo account · referral bindingChina-registered MiMo account · referral binding |
Official source ↗
FreeLLM record ↗ |
| Claude 订阅附送 API 额度 · Max / Team 每月 $100–$500Claude API · Max / Team $100–$500Claude API models on Claude Platform (Console)Claude API models on Claude Platform (Console) |
付费 Claude Max / Team 订阅每月附送 Claude API 额度：Max 5x $100、Max 20x $200、Team 按席位汇总（Standard $20 / Premium $100 每席）上限 $500；不能用于 Claude Code，不结转Paid Claude Max / Team plans include monthly Claude API credits: Max 5x $100, Max 20x $200, Team pooled per seat (Standard $20 / Premium $100) capped at $500; not usable for Claude Code, no rollover |
需要付费 Max 5x / Max 20x / Team 订阅，并把一个 Claude Console 组织关联到订阅（Platform 无需另绑付款方式）Active Max 5x, Max 20x or Team plan (incl. discounted Team plans); new subscribers can claim after 7 days; Free / Pro / Enterprise not eligible |
Official source ↗
FreeLLM record ↗ |
| Requesty · 13 个免费模型共用一把 API KeyRequesty · 13 API KeyNemotron 3 / Gemma 4 / Muse Glimmer / Ling 3.1 Flash 等 13 个 $0 模型Nemotron 3 / Gemma 4 / Muse Glimmer / Ling 3.1 Flash 13 $0 |
官方公开：13 个 $0 模型；新组织合并 200 次/天、20 次/分钟；无需信用卡13 $0 ； 200 / 、20 / ； |
注册账号并建 Key；海外注册/中国大陆网络与年龄验证待实测Key； |
Official source ↗
FreeLLM record ↗ |