FREE AI INDEX / 统一产品界面我的收藏提交资源 ↗

FreeLLM 免费 AI 资源索引FreeLLM Free AI Index / 分类Category

免费额度Free AI quota

整理经过核验的 AI 模型与 API 免费额度和免费入口。Verified recurring free quotas and free access paths for AI models and APIs.

25 个经过核验的资源25 verified resources

怎么选How to choose

先判断它是否真的适合你的使用方式Decide whether the offer actually fits your workload

免费额度适合做原型、学习和低频任务,但要先确认额度是否会恢复,以及输入、输出、缓存和工具调用是否共用同一额度。Free quota is useful for prototypes, learning and light workloads, but first confirm whether it renews and whether input, output, cache and tool calls share the same allowance.

避免误判Avoid this mistake:如果额度只对新用户、指定地域或指定模型有效,我们会在详情页写清,不把局部优惠泛化成整个平台永久免费。When quota applies only to new users, specific regions or selected models, the detail page says so instead of generalizing it into a platform-wide free tier.
FreeLLM 核验方法How FreeLLM verifies

我们优先保留可回到官方来源的事实We keep claims traceable to official sources

详情页尽量把免费方式、额度、有效期、地区、账号要求和官方入口拆开记录;没有证据的结论保持“待核验”,不会为了凑完整而猜测。Detail pages separate the free mechanism, quota, validity, region, account requirements and official source. Claims without evidence remain unverified rather than being guessed.

如果平台规则发生变化,以提供商当前官方页面为准;FreeLLM 的“最后核验”用于告诉你本站什么时候重新检查过这条记录。When provider terms change, the provider's current official page is authoritative. FreeLLM's “last checked” field tells you when we most recently re-verified the record.

FreeLLM 实测FreeLLM test · freellm-text-v2 · 22 题 · 7 维度 · 2026-10-19

本类实测表现Tested in this category

本类 25 个资源中,5 个有 FreeLLM 实测数据;其余 20 个未实测,在下方列表中单独标注。5 of 25 resources here have FreeLLM test data; the other 20 are marked untested below.

资源Resource实测范围Coverage最佳模型 · 首 token · 速度Best model · TTFT · speed明细Details
Groq · free plan rate limits同渠道实测Same-route test测 4 个 · 稳定可用 3 个4 tested · 3 reliablegpt-oss-120b 95 · 202 ms · 481 token/s逐个模型Per model →
OpenRouter · free model variants同渠道实测Same-route test测 2 个 · 稳定可用 1 个2 tested · 1 reliablenemotron-3.5-lightning:free 77 · 160 ms · 86 token/s逐个模型Per model →
SiliconFlow · ¥0 listed models同渠道实测Same-route test测 2 个 · 稳定可用 2 个2 tested · 2 reliableXing4.0-29B 91 · 943 ms · 66 token/s逐个模型Per model →
ModelScope · API-Inference free calls同渠道实测Same-route test测 12 个 · 稳定可用 0 个 · 本账号/本端点不可用 12 个(待复核)12 tested · 0 reliable · 12 unavailable无稳定可用模型No reliable model逐个模型Per model →
GLM-4.7-Flash · BigModel 官方免费 API同渠道实测Same-route test测 2 个 · 稳定可用 1 个2 tested · 1 reliableglm-4-flash-250414 82 · 757 ms · 55 token/s逐个模型Per model →

「同渠道实测」是我们用该平台账号直接调用得到的数字;「仅同款模型参考」只说明同一模型在别的平台的能力得分,不代表本资源的速度和成功率。Same-route numbers come from calling that platform directly; reference rows only show the same model’s capability score elsewhere.

方法:同一套 22 道固定题、temperature=0、全部自动评分;测试节点:GitHub Actions 托管运行器(美国)。数字只代表该时间点、我们的账号和该测试节点,地区与账号不同结果会不同。Method: the same 22 fixed cases, temperature 0, auto-graded; vantage: GitHub Actions (US). Numbers reflect that run, our accounts and that vantage only. 完整报告Full report →

已核验目录Verified directory

当前可用资源Current verified resources

下面每条记录都可以进入详情页查看官方来源、免费条件、有效期和接入要求。Open any record below to review its official source, free-access mechanism, validity and setup requirements.

Google AI Studio · Gemini API free tierGoogle AI Studio · Gemini API 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

Ongoing free tier; limits and terms can changeOngoing free tier; limits and terms can change · Google account; availability may vary by regionGoogle account; availability may vary by region

未实测Not tested

Groq · free plan rate limitsGroq · free plan rate limits

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)

Ongoing free plan; exact limits per model can changeOngoing free plan; exact limits per model can change · Groq account; rate limits apply at organization levelGroq account; rate limits apply at organization level

FreeLLM 实测 4 个模型 · 最高 95 分FreeLLM tested 4 models · best 95

OpenRouter · free model variantsOpenRouter · 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)

Ongoing free usage limits; caps can changeOngoing free usage limits; caps can change · OpenRouter account; daily caps depend on lifetime credit purchasesOpenRouter account; daily caps depend on lifetime credit purchases

FreeLLM 实测 2 个模型 · 最高 77 分FreeLLM tested 2 models · best 77

Mistral · La Plateforme free modeMistral · La Plateforme free mode

Free mode (default): lowest rate limits, intended for evaluation and prototypingFree mode (default): lowest rate limits, intended for evaluation and prototyping

Ongoing default tier; limits can changeOngoing default tier; limits can change · Mistral accountMistral account

未实测Not tested

SiliconFlow · ¥0 listed modelsSiliconFlow · ¥0 listed models◆ 加精◆ Featured

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.

Free listing can change; recheck the price pageFree listing can change; recheck the price page · SiliconFlow account; China registrationSiliconFlow account; China registration

FreeLLM 实测 2 个模型 · 最高 91 分FreeLLM tested 2 models · best 91

ModelScope · API-Inference free callsModelScope · API-Inference free calls◆ 加精◆ Featured

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

Ongoing free service; model list iterates as new open models releaseOngoing free service; model list iterates as new open models release · ModelScope account + Aliyun binding + real-name verificationModelScope account + Aliyun binding + real-name verification

FreeLLM 实测 12 个模型FreeLLM tested 12 models

Agnes AI · 免费/默认多模态 APIAgnes AI · Free/default multimodal API◆ 加精◆ Featured

Free/default API access for selected Agnes models with public rate limitsFree/default API access for selected Agnes models with public rate limits

Ongoing while the official free/default program is available; limits can changeOngoing while the official free/default program is available; limits can change · 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

未实测Not tested

Manus AI · 免费 Agent 计划Manus AI · Agent

新用户注册赠送 1,000 credits;免费计划每天刷新 300 credits1,000 credits; 300 credits

免费计划持续提供;额度、模型和计划内容可能调整; 、 · 浏览器访问;支持 Google、Microsoft、Apple 或邮箱登录; Google、Microsoft、Apple

未实测Not tested

腾讯 WorkBuddy · 办公 AI 智能体Tencent WorkBuddy · office AI agent

个人免费版每月 500 积分,与 CodeBuddy 同账号积分共享,可直接用于 WorkBuddy 办公任务500 , CodeBuddy , WorkBuddy

积分每月刷新;套餐与积分规则以官方价格页为准; · 桌面端(Windows / macOS)为主,另有小程序、移动端与企业版;需登录腾讯账号(Windows / macOS) , 、 ;

未实测Not tested

AMD Radeon Cloud · Token Factory 免费模型 APIAMD Radeon Cloud · Token Factory API

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

Always-on shared tier; AMD may add or retire models at any timeAlways-on shared tier; AMD may add or retire models at any time · Radeon Cloud account verification + rc- API key; shared rate limits applyRadeon Cloud account verification + rc- API key; shared rate limits apply

未实测Not tested

百川 Baichuan-M3 Plus · 医疗机构定向永久免费 APIBaichuan-M3 Plus · API

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

长期计划 · 需申请并通过准入审核Ongoing while the program and approved eligibility remain valid · 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

未实测Not tested

书生·端砚 墨点计划 · 测试期每月免费 TokenToken

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, 。

测试期按月发放 · 结束时间未公布Monthly free allocation during the beta test period; no public end date · 书生·端砚账号 + Token Plan API Key+ Token Plan API Key

未实测Not tested

GLM-4.7-Flash · BigModel 官方免费 APIGLM-4.7-Flash · BigModel API

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.

Official free calls · no fixed public end dateOfficial free calls · no fixed public end date · BigModel account + API KeyBigModel account + API Key

FreeLLM 实测 2 个模型 · 最高 82 分FreeLLM tested 2 models · best 82

Claude 订阅附送 API 额度 · Max / Team 每月 $100–$500Claude API · Max / Team $100–$500

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

每个账单周期发放、周期末过期,不结转;新订阅满 7 天后可领取Refreshes each billing cycle (monthly for annual plans); unused credits expire at cycle end, 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

未实测Not tested

Requesty · 13 个免费模型共用一把 API KeyRequesty · 13 API Key

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 / ;

持续免费档(模型名单、额度及价格可变;官方不承诺永久)( 、 ; ) · 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;

未实测Not tested