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 / 资源详情Offer details

OpenRouter · free model variantsOpenRouter · free model variants

OpenRouter · InternationalOpenRouter

✓ 实测通过✓ Network verified国内外均可用China + global国内 1828ms · 海外 26msChina 1828ms · Global 26ms✓ 接口已验证 · 766ms✓ Endpoint verified · 766ms

这个资源提供什么What this offer provides

OpenRouter's official limits documentation defines free usage limits for model variants whose ID ends in ':free': 20 requests per minute, and 50 requests per day for accounts that purchased less than 10 credits (1000 requests per day at 10 or more credits).OpenRouter's official limits documentation defines free usage limits for model variants whose ID ends in ':free': 20 requests per minute, and 50 requests per day for accounts that purchased less than 10 credits (1000 requests per day at 10 or more credits).

关键信息Key details

免费方式Free method: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)
有效期ValidityOngoing free usage limits; caps can changeOngoing free usage limits; caps can change
访问条件AccessOpenRouter account; daily caps depend on lifetime credit purchasesOpenRouter account; daily caps depend on lifetime credit purchases
最后核验Last checked2026-09-17

快速上手Quick start

Call models with :free variant IDs through the OpenRouter API within the free usage limitsFollow the model-specific steps to register, create a key and connect the model.

✓

注册与接入步骤Registration and access steps

准备条件Prerequisites

  • OpenRouter account
  • API key from the OpenRouter dashboard
  1. Create an OpenRouter account and generate an API keyCreate an OpenRouter account and generate an API key
  2. Send chat completion requests to model IDs ending in :free and observe the published per-day request capsSend chat completion requests to model IDs ending in :free and observe the published per-day request caps
4

发送第一条请求Send first request

使用下方命令开始调用。Use the command below to make your first request.

可复制命令Copyable command
export OPENROUTER_API_KEY="你的 Key"
curl https://openrouter.ai/api/v1/chat/completions --header "Authorization: Bearer $OPENROUTER_API_KEY" --header 'Content-Type: application/json' --data '{"model":"inclusionai/ling-3.0-flash-fin:free","messages":[{"role":"user","content":"Hello"}]}'

详细操作步骤Detailed operation paths

每条路径都拆成前置条件、步骤、可复制命令和验证动作;免费条件仍以官方页面实时状态为准。curl 命令按 macOS / Linux shell 编写,Windows 用户可使用同页的 Python 或 Node.js 示例。Each path includes prerequisites, steps, copyable commands and a validation action; free terms still follow the provider's live official policy. The curl examples use a macOS/Linux shell; Windows users can run the Python or Node.js examples on this page.

调用免费模型

api

EndpointEndpoint
https://openrouter.ai/api/v1
认证Authentication
Authorization: Bearer API key
额度与限制Limits

调用额度与限流按提供商当前控制台和官方文档为准;免费条件可能按模型、账号和地区变化。

前置条件Prerequisites

  • 已注册 OpenRouterOpenRouter
  • 已创建 API keyAPI key
  • 已从模型页确认当前免费模型 IDID

操作步骤Steps

  1. 创建 API key

    进入 OpenRouter Keys 页面创建密钥,并将它放入 OPENROUTER_API_KEY 环境变量。

  2. 选择免费模型

    在官方 Models 页面筛选价格为 0 的模型;免费标记和可用性会变化,不要长期写死目录中的模型。

    操作Action:打开 https://openrouter.ai/models?max_price=0

  3. 发送 OpenAI 兼容请求

    将 model 替换成模型页显示的完整 ID。

    命令已在上方「快速上手」给出。The copyable command already appears in Quick start above.

    预期结果Expected:响应包含 choices[0].message.content。

  4. Python SDK 示例

    安装 OpenAI SDK;在运行环境设置 OPENROUTER_API_KEY,示例代码通过环境变量读取密钥。

    可复制命令Copyable command
    python -m pip install -U openai
    
    import os
    from openai import OpenAI
    
    client = OpenAI(
        api_key=os.environ["OPENROUTER_API_KEY"],
        base_url="https://openrouter.ai/api/v1",
    )
    response = client.chat.completions.create(
        model="openai/gpt-oss-20b:free",
        messages=[{"role": "user", "content": "用一句话解释 API 是什么。"}],
    )
    print(response.choices[0].message.content)

    预期结果Expected:终端打印模型生成的回复;再到提供商控制台确认请求状态和额度扣减。

  5. Node.js SDK 示例

    安装 OpenAI SDK;在运行环境设置 OPENROUTER_API_KEY,示例代码通过环境变量读取密钥。

    可复制命令Copyable command
    npm install openai
    
    import OpenAI from "openai";
    
    const client = new OpenAI({
      apiKey: process.env.OPENROUTER_API_KEY,
      baseURL: "https://openrouter.ai/api/v1",
    });
    const response = await client.chat.completions.create({
      model: "openai/gpt-oss-20b:free",
      messages: [{ role: "user", content: "用一句话解释 API 是什么。" }],
    });
    console.log(response.choices[0].message.content);

    预期结果Expected:Node.js 打印模型生成的回复;再到提供商控制台确认请求状态和额度扣减。

验证动作Validation

GET /api/v1/models 返回 HTTP 200;随后聊天请求返回 choices 数组。

常见问题Common issues
  • 401:检查 API Key、环境变量和认证 Header。401: API Key、 Header。
  • 404:从官方模型清单复制当前模型 ID,并确认 endpoint 与账号地域一致。404: ID, endpoint 。
  • 429:查看当前限流和配额重置说明,降低请求频率后再重试。429: , 。

官方来源Official sources

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

FreeLLM 模型实测FreeLLM model test

我们用自己的账号直接调用该平台得到的结果(同渠道实测)。Results from calling this platform directly with our own key.

模型Model综合分Score成功率Success首 token p50TTFT p50生成速度Tokens/s报告Report
nemotron-3.5-lightning:freenvidia/nemotron-3.5-lightning:free77100%160 msp95 16975 ms86 token/s报告Report →
gemma-4-31b-it:freegoogle/gemma-4-31b-it:free27 受失败影响affected by failures27%815 msp95 969 ms41 token/s报告Report →

nemotron-3.5-lightning:free

nvidia/nemotron-3.5-lightning:free · OpenRouter · 默认思考模式thinks by default
77分pts
综合分Score77 / 100
成功率Success rate100%(22/22 题有回答)
首 token p50 / p95TTFT p50 / p95160 ms / 16975 ms
生成速度(中位)Tokens/s (median)86 token/s
  • 中文知识Chinese knowledge67
  • 推理数学Reasoning & math100
  • 代码Coding100
  • 指令格式Format following67
  • 长文本Long context100
  • 工具调用Tool calling67
  • 防幻觉Anti-hallucination33
首字极快Very fast first token推理数学满分Full marks: Reasoning & math代码满分Full marks: Coding长文本满分Full marks: Long context默认长思考,响应慢Thinks by default, slower容易编造Tends to fabricate

推理、代码、长文本全对,但常用英文作答(城市名答成 Nanchang、JSON 输出英文),一题输出陷入「Here is」无限重复。

适合Good for
英文或中英混合的推理与代码任务
不建议Avoid for
要求纯中文输出的产品、实时交互(最慢一题首字约 28 秒)

gemma-4-31b-it:free

google/gemma-4-31b-it:free · OpenRouter
27%请求成功率;限流/失败较多,27 分受失败请求影响request success; score 27 is dragged down by failures
综合分Score27 / 100
成功率Success rate27%(6/22 题有回答)
首 token p50 / p95TTFT p50 / p95815 ms / 969 ms
生成速度(中位)Tokens/s (median)41 token/s
  • 中文知识Chinese knowledge33
  • 推理数学Reasoning & math40
  • 代码Coding67
  • 指令格式Format following33
  • 长文本Long context0
  • 工具调用Tool calling0
  • 防幻觉Anti-hallucination0
中文知识弱Weak: Chinese knowledge指令格式弱Weak: Format following长文本弱Weak: Long context工具调用弱Weak tool calling

本轮 22 题全部被 OpenRouter 上游限流(429),是免费额度限制,不代表模型能力,待下一轮复测。

适合Good for
暂无实测结论
不建议Avoid for
依赖 OpenRouter 免费档稳定调用的场景

查看该平台全部 2 个实测模型(含不可用端点)All 2 tested models on this platform →

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 →

官方来源Official sources

Official docs table: 'Credits purchased (all time) | Requests per minute | Requests per day; Less than 10 | 20 | 50; At least 10 | 20 | 1000', applied to 'a free model variant (with an ID ending in :free)'. The official GET /api/v1/models catalog currently lists 18 model IDs ending in :free (e.g. inclusionai/ling-3.0-flash-fin:free).Official docs table: 'Credits purchased (all time) | Requests per minute | Requests per day; Less than 10 | 20 | 50; At least 10 | 20 | 1000', applied to 'a free model variant (with an ID ending in :free)'. The official GET /api/v1/models catalog currently lists 18 model IDs ending in :free (e.g. inclusionai/ling-3.0-flash-fin:free).

相关资源Related resources