Skip to content

Latest commit

 

History

History
335 lines (276 loc) · 32.2 KB

File metadata and controls

335 lines (276 loc) · 32.2 KB

awesome-free-llm-apis

243+ 免費大模型 API,來自 28 個提供商 — 一站式發現、對比、配置免費模型。

🌐 訪問 freellm.net瀏覽模型 · 線上體驗 · 配置生成器 · API 金鑰

🔄 資料每日從 freellm.net 自動更新 — 最後更新: 2026-06-18

🌐 English · 简体中文 · 繁體中文 · 日本語 · 한국어


為什麼需要這個專案

找一個免費 LLM API,不應該翻十幾個 GitHub README、註冊五個不同平台、或者猜測哪些模型還有免費額度。

這個倉庫是一個結構化、機器可讀的免費 LLM API 目錄 — 包含速率限制、上下文視窗、一鍵配置程式碼片段和直達 API 金鑰連結。每日更新。

為什麼選擇這個倉庫 + freellm.net

  • 始終最新 — 資料每日自動化監控更新,不是兩年前的靜態列表
  • 信用卡透明 — 清晰標註哪些提供商需要信用卡、手機驗證,哪些完全無需
  • 一鍵配置 — 即用程式碼片段覆蓋 Claude Code、Cursor、Codex、Aider 等 10+ 工具
  • 並排對比 — 即時對比各提供商的上下文視窗、速率限制和多模態能力

三步上手

  1. 選一個提供商 — 看下方 Provider Directory。推薦從 Groq 開始(免信用卡,30 RPM 免費)。
  2. 獲取 API Key — 點下方 Quick Reference 裡任意一個 [Get Key →] 連結,註冊(大多只需信箱),複製 key。不到 1 分鐘。
  3. 填入程式碼 — 複製 base URL + model ID,貼到下邊的 Quick Start 範例。

配置具體工具?Claude Code · Cursor · Codex · OpenHuman · OpenCode · OpenClaw — 一鍵配置在 freellm.net/config/

快速上手 — 30 秒接入免費 API

沒用過 API? 最簡單的方式:打開 console.groq.com/keys,用信箱註冊(無需信用卡),複製免費 key,貼到下面任意範例。不到一分鐘就能跑起來。

以下所有提供商都暴露了 OpenAI 相容介面。任何接受 baseURL + apiKey 的工具都能直接用 — 替換 base URL 和 key 即可。

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://api.groq.com/openai/v1",  # 免費,無需信用卡
    api_key="GROQ_API_KEY",                     # 獲取: console.groq.com/keys
)

response = client.chat.completions.create(
    model="llama-3.3-70b-versatile",            # 模型列表見下方 Best Models 表
    messages=[{"role": "user", "content": "你好!"}],
)
print(response.choices[0].message.content)
# Groq 免費額度: 30 RPM, 14,400 RPD — 個人使用綽綽有餘

Codex CLI

export OPENAI_BASE_URL="https://api.groq.com/openai/v1"
export OPENAI_API_KEY="your-groq-key"          # 獲取: console.groq.com/keys
codex --model "llama-3.3-70b-versatile"

Cursor

Settings → Models → Add Model
  Model name: llama-3.3-70b-versatile
  Base URL: https://api.groq.com/openai/v1
  API key: your-groq-key                       # 獲取: console.groq.com/keys

Claude Code

# Claude Code 需要 Anthropic 相容介面 — 使用 OpenRouter
export ANTHROPIC_BASE_URL="https://openrouter.ai/api"
export ANTHROPIC_AUTH_TOKEN="sk-or-v1-your-key"  # openrouter.ai/keys
export ANTHROPIC_API_KEY=""                       # 必須為空
# 注意: OpenRouter 的 Anthropic 模型需儲值 $10(一次性)

其他工具?

大多數 AI 開發工具都接受自訂 API 端點 — 拿到免費 key 後:

更多即用配置在 freellm.net/config/

所有提供商的 Base URL 和 API Key 連結 見下方 Quick Reference


提供商目錄 & 熱門免費模型

Provider Free Models Credit Card? Max Context Modalities Get API Key
NVIDIA NIM 75 Phone verification 1M image, text
Groq 14 No 262K text
Cloudflare Workers AI 13 No 10M code, image, text
GitHub Models 13 No 1M text
OVHcloud AI Endpoints 12 Registration 262K code, image, text
Mistral AI 9 No 256K code, image, text
Cohere 6 No 256K text
Cerebras 6 No 131K text
LLM7.io 6 No 131K code, text
Ollama Cloud 6 Registration 262K code, text
Aion Labs 5 Registration 131K text
Google Gemini 5 No 2M text
Hugging Face 5 No 131K text
Kilo Code 5 No 262K code, text
Alibaba Cloud Model Studio 5 Registration 1M code, image, text
SambaNova 4 Registration 128K text
Z AI (Zhipu AI) 3 No 200K text
ModelScope 3 Registration 131K text
SiliconFlow 3 Registration 131K text
xAI 3 Registration 2M text
AI21 Labs 2 Registration 256K text
DeepSeek 2 Registration 128K text
Nscale 2 Registration 128K text
Chutes.ai 2 Registration 131K text
Glhf.chat 2 Registration 131K text
Grok (xAI) 2 Registration 131K text
Nebius 1 Registration 128K text
Provider Free Models Credit Model Max Context Modalities Get API Key
OpenRouter 29 Free tier + $10 topup → 1K RPD 1M audio, code, embeddings, image, reasoning, rerank, text

快速參考 — Base URL 與 API Key

Provider Base URL Get API Key Credit Card?
NVIDIA NIM https://integrate.api.nvidia.com/v1 Get Key → Phone verification
OpenRouter https://openrouter.ai/api/v1 Get Key → Registration
Groq https://api.groq.com/openai/v1 Get Key → No
Cloudflare Workers AI https://api.cloudflare.com/client/v4/accounts/{account_id}/ai/run Get Key → No
GitHub Models https://models.github.ai/inference Get Key → No
OVHcloud AI Endpoints https://oai.endpoints.kepler.ai.cloud.ovh.net/v1 Get Key → Registration
Mistral AI https://api.mistral.ai/v1 Get Key → No
Cohere https://api.cohere.com/v2 Get Key → No
Cerebras https://api.cerebras.ai/v1 Get Key → No
LLM7.io https://api.llm7.io/v1 Get Key → No
Ollama Cloud https://api.ollama.com Get Key → Registration
Aion Labs https://api.aionlabs.ai/v1 Get Key → Registration
Google Gemini https://generativelanguage.googleapis.com/v1beta Get Key → No
Hugging Face https://router.huggingface.co/v1 Get Key → No
Kilo Code https://api.kilo.ai/api/gateway Get Key → No
Alibaba Cloud Model Studio https://dashscope-intl.aliyuncs.com/compatible-mode/v1 Get Key → Registration
SambaNova https://api.sambanova.ai/v1 Get Key → Registration
Z AI (Zhipu AI) https://open.bigmodel.cn/api/paas/v4 Get Key → No
ModelScope https://api-inference.modelscope.cn/v1 Get Key → Registration
SiliconFlow https://api.siliconflow.cn/v1 Get Key → Registration
xAI https://api.x.ai/v1 Get Key → Registration
AI21 Labs https://api.ai21.com/studio/v1 Get Key → Registration
DeepSeek https://api.deepseek.com/v1 Get Key → Registration
Nscale https://inference.api.nscale.com/v1 Get Key → Registration
Chutes.ai https://api.chutes.ai/v1 Get Key → Registration
Glhf.chat https://glhf.chat/api/openai/v1 Get Key → Registration
Grok (xAI) https://api.x.ai/v1 Get Key → Registration
Nebius https://api.studio.nebius.com/v1 Get Key → Registration

各提供商最佳免費模型

Provider Best Free Model Model ID Max Context Rate Limit
NVIDIA NIM z-ai/glm-5.1 z-ai/glm-5.1 202K Up to 40 RPM
moonshotai/kimi-k2.6 moonshotai/kimi-k2.6 262K Up to 40 RPM
qwen/qwen3.5-397b-a17b qwen/qwen3.5-397b-a17b 256K Up to 40 RPM
OpenRouter Owl Alpha openrouter/owl-alpha 1M See provider
NVIDIA: Nemotron 3 Ultra (free) nvidia/nemotron-3-ultra-550b-a55b:free 1M See provider
Nex AGI: Nex-N2-Pro (free) nex-agi/nex-n2-pro:free 262K See provider
Groq llama-3.3-70b-versatile llama-3-3-70b-versatile 131K 30 RPM, 1,000 RPD
llama-3.1-8b-instant llama-3-1-8b-instant 131K 30 RPM, 1,000 RPD
llama-4-scout-17b-16e-instruct llama-4-scout-17b-16e-instruct 131K 30 RPM, 1,000 RPD
Cloudflare Workers AI @cf/meta/llama-3.3-70b-instruct-fp8-fast cf-meta-llama-3-3-70b-instruct-fp8-fast 131K 10K neurons/day (shared)
@cf/meta/llama-4-scout-17b-16e-instruct cf-meta-llama-4-scout-17b-16e-instruct 10M 10K neurons/day (shared)
@cf/openai/gpt-oss-120b cf-openai-gpt-oss-120b 128K 10K neurons/day (shared)
GitHub Models gpt-5 gpt-5 200K 10 RPM, 50 RPD
gpt-4.1 gpt-4-1 1M 10 RPM, 50 RPD
gpt-4.1-mini gpt-4-1-mini 1M 15 RPM, 150 RPD
OVHcloud AI Endpoints Qwen3.5-397B-A17B qwen3-5-397b-a17b 131K 2 RPM (anonymous)
gpt-oss-20b gpt-oss-20b 128K 2 RPM (anonymous)
Meta-Llama-3_3-70B-Instruct meta-llama-3-3-70b-instruct 131K 2 RPM (anonymous)
Mistral AI Mistral Medium 3.5 (128B) mistral-medium-3-5-128b 256K ~1 RPS, 500K TPM
Mistral Small 4 mistral-small-4 256K ~1 RPS, 500K TPM
Mistral Large 3 mistral-large-3 256K ~1 RPS, 500K TPM
Cohere Command A+ (218B) command-a-218b 128K 20 RPM
Command A (111B) command-a-111b 256K 20 RPM
Command R+ command-r 128K 20 RPM
Cerebras gpt-oss-120b gpt-oss-120b 128K 30 RPM, 14,400 RPD, 1M ..
zai-glm-4.7 zai-glm-4-7 128K 10 RPM, 100 RPD, 1M TPD
llama-3.3-70b llama-3-3-70b 128K 30 RPM, 14,400 RPD, 1M ..
LLM7.io deepseek-r1-0528 deepseek-r1-0528 131K 30 RPM (120 with token)
deepseek-v3-0324 deepseek-v3-0324 131K 30 RPM (120 with token)
gemini-2.5-flash-lite gemini-2-5-flash-lite 131K 30 RPM (120 with token)
Ollama Cloud gpt-oss:120b-cloud gpt-oss-120b-cloud 128K Session/weekly limits (..
deepseek-v3.1:671b-cloud deepseek-v3-1-671b-cloud 128K Session/weekly limits (..
qwen3-coder:480b-cloud qwen3-coder-480b-cloud 128K Session/weekly limits (..
Aion Labs Aion 2.5 aion-2-5 128K 15 RPM, 20K TPD
Aion 2.0 aion-2-0 128K 15 RPM, 20K TPD
Aion-RP 1.0 (8B) aion-rp-1-0-8b 32K 15 RPM, 20K TPD
Google Gemini Gemini 3.5 Flash gemini-3-5-flash 1M 15 RPM, 1,500 RPD
Gemini 3.1 Flash-Lite gemini-3-1-flash-lite 1M 30 RPM, 1,500 RPD
Gemini 2.5 Flash gemini-2-5-flash 1M 15 RPM, 1,500 RPD
Hugging Face Meta-Llama-3.1-8B-Instruct meta-llama-3-1-8b-instruct 128K Credit-metered
Mistral-7B-Instruct-v0.3 mistral-7b-instruct-v0-3 32K Credit-metered
Mixtral-8x7B-Instruct-v0.1 mixtral-8x7b-instruct-v0-1 32K Credit-metered
Kilo Code x-ai/grok-code-fast-1:free x-ai-grok-code-fast-1-free 256K ~200 req/hr
minimax/minimax-m2.5:free minimax-minimax-m2-5-free 196K ~200 req/hr
bytedance-seed/dola-seed-2.0-pro:free bytedance-seed-dola-seed-2-0-pro-free 131K ~200 req/hr
Alibaba Cloud Model Studio Qwen3-Max qwen3-max 128K Tiered by region
Qwen3-Plus qwen3-plus 1M Tiered by region
Qwen3-VL-Plus qwen3-vl-plus 128K Tiered by region
SambaNova DeepSeek-V3.1 deepseek-v3-1 128K 20 RPM, 20 RPD, 200K TPD
DeepSeek-V3.2 (Preview) deepseek-v3-2-preview 128K 20 RPM, 20 RPD, 200K TPD
MiniMax-M2.7 minimax-m2-7 128K 20 RPM, 20 RPD, 200K TPD
Z AI (Zhipu AI) GLM-4.7-Flash glm-4-7-flash 200K 1 concurrent request
GLM-4.6V-Flash glm-4-6v-flash 128K 1 concurrent request
GLM-4.5-Flash glm-4-5-flash 128K 1 concurrent request
ModelScope Qwen/Qwen3.5-35B-A3B qwen-qwen3-5-35b-a3b 131K 2,000 RPD total; <=500 ..
Qwen/Qwen3.5-27B qwen-qwen3-5-27b 131K 2,000 RPD total; <=500 ..
Qwen/Qwen-Image qwen-qwen-image 131K 2,000 RPD total; model/..
SiliconFlow deepseek-ai/DeepSeek-R1-Distill-Qwen-7B deepseek-ai-deepseek-r1-distill-qwen-7b 131K 30 RPM, 60K TPM
Abbreviation abbreviation 131K See provider
deepseek-ai/DeepSeek-OCR deepseek-ai-deepseek-ocr 131K 30 RPM, 60K TPM
xAI grok-4.3 grok-4-3 1M Credit-based
grok-4.1-fast grok-4-1-fast 2M Credit-based
grok-3-mini grok-3-mini 131K Credit-based
AI21 Labs Jamba Large 1.7 jamba-large-1-7 256K 200 RPM, 10 RPS
Jamba Mini 2 jamba-mini-2 256K 200 RPM, 10 RPS
DeepSeek deepseek-chat (V3.2) deepseek-chat-v3-2 128K Dynamic
deepseek-reasoner (R1) deepseek-reasoner-r1 128K Dynamic
Nscale Llama-3.3-70B-Instruct llama-3-3-70b-instruct 128K Fair-use
DeepSeek-R1-Distill-Llama-70B deepseek-r1-distill-llama-70b 128K Fair-use
Chutes.ai DeepSeek-R1 deepseek-ai/DeepSeek-R1 131K Community-powered, no h..
Llama 3.1 70B meta-llama/Meta-Llama-3.1-70B-Instruct 131K Community-powered, no h..
Glhf.chat Llama 3.1 70B meta-llama/Meta-Llama-3.1-70B-Instruct 131K Unlimited for free models
Mixtral 8x7B mistralai/Mixtral-8x7B-Instruct-v0.1 32K Unlimited for free models
Grok (xAI) Grok-2 grok-2 131K $25/month free credits,..
Grok-2 Mini grok-2-mini 131K $25/month free credits,..
Nebius Qwen3-235B-A22B qwen3-235b-a22b 128K Tier-based

热门免费模型

Model Provider Context Weekly Usage
Owl Alpha OpenRouter 1M 2448B tokens
z-ai/glm-5.1 NVIDIA NIM 202K 957B tokens
NVIDIA: Nemotron 3 Ultra (free) OpenRouter 1M 762B tokens
Nex AGI: Nex-N2-Pro (free) OpenRouter 262K 710B tokens
Poolside: Laguna M.1 (free) OpenRouter 262K 597B tokens
moonshotai/kimi-k2.6 NVIDIA NIM 262K 491B tokens
NVIDIA: Nemotron 3 Super (free) OpenRouter 1M 363B tokens
OpenAI: gpt-oss-120b (free) OpenRouter 131K 206B tokens
Poolside: Laguna XS.2 (free) OpenRouter 262K 95B tokens
OpenAI: gpt-oss-20b (free) OpenRouter 131K 58B tokens

參與貢獻

歡迎貢獻!

  • 補充缺失的免費模型 — 提交 issue 或 PR
  • 修正不準確的資料 — 速率限制會變、提供商會升級,歡迎 PR
  • 添加配置程式碼片段 — 有你常用工具的配置方案?添加到 code-examples/

收錄標準

模型列入本列表需滿足:

  1. 提供商明確提供免費層(不僅僅是試用額度)
  2. API 公開可訪問(無需候補、封閉內測、或逆向工程)
  3. 試用額度:明確標註且最低 $1 額度

連結

授權

MIT © open-free-llm-api


資料每日自動更新 · 最後更新: 2026-06-18