🛠
AI工具
AI编码优惠
基于 Python · 开源免费，本地部署，数据完全自主可控
英文名：ai-coding-deals
8.0AI 综合评分
aiai-codingai-toolspython
✦ AI Skill Hub 推荐
AI Skill Hub 强烈推荐：AI编码优惠 是一款优质的AI工具。AI 综合评分 8.0 分，在同类工具中表现稳健。如果你正在寻找可靠的AI工具解决方案，这是一个值得深入了解的选择。
📚 深度解析
AI编码优惠 是一款基于 Python 的开源工具，在 GitHub 上收获 0k+ Star，是ai、ai-coding、ai-tools、python领域中的优质开源项目。开源工具的最大优势在于代码完全透明，你可以审计每一行代码的安全性，也可以根据自身需求进行二次开发和定制。
**为什么要使用开源工具而非商业 SaaS？**
对于个人开发者和有隐私需求的用户，本地部署的开源工具意味着数据不离本机，不受第三方服务商的数据政策约束。同时，开源工具通常没有使用次数限制和月度费用，一次安装即可长期使用，对于高频使用场景的总拥有成本（TCO）远低于订阅制商业工具。
**安装与环境准备**
AI编码优惠 依赖 Python 运行环境。建议通过 pyenv（Python）或 nvm（Node.js）管理 Python 版本，避免全局环境污染。对于新手用户，推荐先创建虚拟环境（python -m venv venv && source venv/bin/activate），再安装依赖，这样即使出现问题也可以随时删除虚拟环境重新开始，不影响系统稳定性。
**社区与维护**
GitHub Issue 和 Discussion 是获取帮助的最快渠道。在提问前建议先检查 Closed Issues（已关闭的问题），大多数常见问题都已有解答。遇到 Bug 时，提供 pip list 的输出、完整错误堆栈和最小可复现示例，能显著提高开发者响应速度。AI Skill Hub 将持续追踪 AI编码优惠 的版本更新，及时通知重要功能变化。
**为什么要使用开源工具而非商业 SaaS？**
对于个人开发者和有隐私需求的用户，本地部署的开源工具意味着数据不离本机，不受第三方服务商的数据政策约束。同时，开源工具通常没有使用次数限制和月度费用，一次安装即可长期使用，对于高频使用场景的总拥有成本（TCO）远低于订阅制商业工具。
**安装与环境准备**
AI编码优惠 依赖 Python 运行环境。建议通过 pyenv（Python）或 nvm（Node.js）管理 Python 版本，避免全局环境污染。对于新手用户，推荐先创建虚拟环境（python -m venv venv && source venv/bin/activate），再安装依赖，这样即使出现问题也可以随时删除虚拟环境重新开始，不影响系统稳定性。
**社区与维护**
GitHub Issue 和 Discussion 是获取帮助的最快渠道。在提问前建议先检查 Closed Issues（已关闭的问题），大多数常见问题都已有解答。遇到 Bug 时，提供 pip list 的输出、完整错误堆栈和最小可复现示例，能显著提高开发者响应速度。AI Skill Hub 将持续追踪 AI编码优惠 的版本更新，及时通知重要功能变化。
📋 工具概览
AI编码优惠 是一款基于 Python 开发的开源工具，专注于 ai、ai-coding、ai-tools 等核心功能。作为 GitHub 开源项目，它拥有活跃的社区支持和持续的版本迭代，代码完全透明可审计，支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流，都能提供稳定可靠的解决方案。
GitHub Stars
⭐ 45
开发语言
Python
支持平台
Windows / macOS / Linux
维护状态
轻量级项目，按需更新
开源协议
MIT
AI 综合评分
8.0 分
工具类型
AI工具
Forks
1
📖 中文文档
以下内容由 AI Skill Hub 根据项目信息自动整理，如需查看完整原始文档请访问底部「原始来源」。
AI编码优惠 是一款基于 Python 开发的开源工具，专注于 ai、ai-coding、ai-tools 等核心功能。作为 GitHub 开源项目，它拥有活跃的社区支持和持续的版本迭代，代码完全透明可审计，支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流，都能提供稳定可靠的解决方案。
📌 核心特色
- 开源免费，支持本地部署，数据完全自主可控
- 活跃的 GitHub 开源社区，持续迭代更新
- 提供详细文档和使用示例，新手友好
- 支持自定义配置，灵活适配不同使用环境
- 可作为基础组件集成进现有技术栈或进行二次开发
🎯 主要使用场景
- 本地部署运行，保护数据隐私，满足合规要求
- 自定义集成到现有系统，扩展技术栈能力
- 作为开源基础组件进行商业化二次开发
以下安装命令基于项目开发语言和类型自动生成，实际以官方 README 为准。
安装命令
# 方式一：pip 安装（推荐）
pip install ai-coding-deals
# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install ai-coding-deals
# 方式三：从源码安装（获取最新功能）
git clone https://github.com/codertesla/ai-coding-deals
cd ai-coding-deals
pip install -e .
# 验证安装
python -c "import ai_coding_deals; print('安装成功')"
📋 安装步骤说明
- 访问 GitHub 仓库页面
- 按照 README 文档完成依赖安装
- 根据系统环境完成初始化配置
- 参考官方示例或文档开始使用
- 遇到问题可在 GitHub Issues 中查找解答
以下用法示例由 AI Skill Hub 整理，涵盖最常见的使用场景。
常用命令 / 代码示例
# 命令行使用
ai-coding-deals --help
# 基本用法
ai-coding-deals input_file -o output_file
# Python 代码中调用
import ai_coding_deals
# 示例
result = ai_coding_deals.process("input")
print(result)
以下配置示例基于典型使用场景生成，具体参数请参照官方文档调整。
配置示例
# ai-coding-deals 配置文件示例（config.yml） app: name: "ai-coding-deals" debug: false log_level: "INFO" # 运行时指定配置文件 ai-coding-deals --config config.yml # 或通过环境变量配置 export AI_CODING_DEALS_API_KEY="your-key" export AI_CODING_DEALS_OUTPUT_DIR="./output"
以下内容由系统直接从 GitHub README 解析整理，保留代码块、表格与列表结构。
🔥 Featured Deals & Cost-Savers
Curated by recommendation priority (higher rows first). Expired, paused-signup, or niche-only offers were removed from this table — they remain in the category sections below. (Refreshed August 2026 — confirm on the official vendor page before purchasing.)
[!TIP] DeepSeek-V4-Flash-0731 (official public beta, 2026-07-31): Same architecture with a post-training upgrade — substantially stronger agentic / coding / tool-calling performance, plus native Responses API support (Codex-ready). Keep calling deepseek-v4-flash (DeepSeek API Docs). V4-Pro GA is still pending. Two best-value channels for the latest V4 Flash today: ① Freebuff (ad-supported, $0); ② OpenCode Go ($5 first month / $10 after, ~$60/mo Flash-equivalent usage — unit prices match official, so fully used cost is ~1/6 of pay-as-you-go).
| Tool | The deal | Why it's worth it |
|---|---|---|
| **OpenCode Go** ✅ *verified* | **$5 free credit** for both parties via [referral link](https://opencode.ai/go?ref=J9E8732NMP); **Go plan: $5 first month, then $10/mo** (~**$60/mo** equivalent usage; $12/5-hour, $30/week). **DeepSeek V4 Flash** gets the full **$60** allowance at official unit prices ($0.14 / $0.28 / $0.0028 cache-hit per 1M tokens) — ~**1/6** effective cost vs official pay-as-you-go when fully used; V4 Pro is capped at **$15**, so heavy users should prefer Flash | **Cheapest paid path to latest V4 Flash vs official metering**; API key works with third-party agent clients |
| **Cursor** ✅ *verified* | **50% off your first month** for new accounts via any referral link — Pro for **$10** (first month), with proportional discounts on higher tiers. Referral links have usage limits; search for a fresh link (e.g., search "Cursor referral link"). **Alipay accepted** — remember to cancel auto-renew if you do not plan to continue. | Leading Agentic IDE; verified June 2026 — the lowest-cost entry to full Pro features |
| **Freebuff** ✅ *verified* | **100% free** ad-supported coding agent (CLI / Web / Chat / Desktop) at [freebuff.com](https://freebuff.com/) — no API key or subscription; includes **DeepSeek V4 Flash / Pro** (incl. latest **Flash-0731**) plus Kimi, MiniMax, and more; [Web referral link](https://freebuff.com/web/?ref=ref-b3046190-1d77-44c1-915c-b71b4f36ae5e) | **Best $0 path to the latest DeepSeek V4** — simplest zero-budget trial of V4 agent capabilities |
| **Kimi**（Moonshot） | **Limited-time free membership lottery** via [invite link](https://kimi-bot.com/activities/zh-cn/viral-referral/share?scenario=invite&from=share_poster&invitation_code=ZCDJXZ) — prizes of **3 / 7 / 15 / 30 / up to 365 days** (minimum 3 days); flagship **Kimi K3** released **2026-07-17**, ranking **TOP 3** on multiple benchmarks; paid plans at [membership pricing](https://www.kimi.com/membership/pricing?from=kfc_membership_upgrade_price&track_id=fcbd4630-4439-498a-b934-d6cee41da4f1) (CN tiers from Andante **¥49/mo**, incl. Kimi Code / Agent) | Lottery odds are uncertain; **for a more reliable free K3 trial, prefer WorkBuddy first** (first-month credits usually suffice) |
| **Tencent Cloud AI Model Square** ✅ *verified* | New users can visit the [Model Square](https://curl.qcloud.com/PN0cvBnI) and click "**New User Benefits Free Experience**" (top-right corner) to claim — **all language models** and **all multimodal understanding models** each receive **1M free trial tokens**, valid for **90 days** from claim (incl. **HY3**, GLM-5.2, DeepSeek V4 Pro, Kimi K2.7 Code, etc.); Hunyuan **HY3** also offers a dedicated **Hy Token Plan** starting at **¥28/month for 35M tokens** | Best low/zero-cost API base for regional BYOK clients — one key covers multiple flagship models |
| **OpenAI Codex** | **Included in ChatGPT Free** — accessible via Web, CLI, and **desktop applications (macOS/Windows)** with no credit card required | Broadest zero-cost entry; any ChatGPT Free account can start immediately |
| **Google Antigravity / Gemini** | Generous **free tier**; AI Pro subscription available at **$19.99/month** | Generous free tier; lowest-friction way to try multi-agent / fast Gemini workflows |
| **Grok / SuperGrok**（xAI）✅ *verified* | **7-day SuperGrok free trial** via [grok.com/plans](https://grok.com/plans) — unlocks **Grok 4.5** and **Grok Build** (xAI terminal coding agent) at full paid limits; **credit card required**; cancel before the trial ends or you are billed **$30/month**. Likely limited to accounts that have **never subscribed** before; promo end date unclear | One free week of frontier model + terminal agent — cancel before the trial ends |
| **Cline (ClinePass)** | **$1.99/month launch promo** (standard **$9.99/month**) when signing up via npm i -g cline — now includes flagship **Kimi K3** at official **~5× discounted access**; also bundled GLM-5.2, Kimi K2.7 Code, DeepSeek V4, Qwen3.7, MiniMax, MiMo, and more; available on Cline CLI & IDE | One of the lower-cost paid paths to **Kimi K3** (especially while Kimi capacity is constrained); regional new users wanting a free trial should prefer WorkBuddy first; promo end date TBD — verify at [cline.bot](https://cline.bot) or [x.com/cline](https://x.com/cline) |
| **CodeBuddy / WorkBuddy**（Tencent, regional）✅ *verified* | Sign up via [WorkBuddy invite link](https://www.workbuddy.cn/events/invite?inviteCode=fmfw2whmh) (**WeChat login**): invitees receive a **2,000-credit signup bonus** (valid **1 month**); completing onboarding tasks grants additional credits for a **total of up to 3,600 credits** — roughly equivalent to one free month of the Standard plan (**¥99/mo**); **on first paid purchase, both parties receive 500 credits** (invitee bonus once only, valid **6 months**), and invitees still qualify for other promos; plus **HY3 model free for all users** (0 credits, **2026-07-06 through 2026-07-22**); now includes flagship **Kimi K3** (~**1.62×** credit multiplier) | Strongest regional new-user credit pack — roughly one free Standard month; **prefer this first if you want to try Kimi K3 for free** — first-month credits are usually enough for a solid trial |
💡 Tip: Most subscription-based AI coding tools offer ~15%–20% savings when billed annually.
Have an active referral link or a new promotion to share? Please submit a Pull Request — see Contributing.
AI Coding Cost Optimization & Selection Guide 💰
A comprehensive guide to AI coding tool selection and cost optimization — tracking free tiers, commercial discounts, referral credits, and open-source alternatives in one place.
AI coding tool pricing and quotas evolve rapidly. This repository tracks free tiers, commercial discounts, referral credits, and highly cost-effective deployment strategies to help developers and technical teams optimize their AI spend. Maintained by the community and verified against official vendor documentation.
[!IMPORTANT] Disclaimer: All information is aggregated from public sources and is subject to change. Always verify pricing, quotas, and terms on the vendor's official website before making purchasing decisions. Spot an error or outdated entry? Please open an issue or submit a Pull Request. Last comprehensive verification: June 2026.
🏗️ Agentic IDEs (AI-Native Integrated Development Environments)
Complete integrated development environments featuring deep agentic workflows — allowing the AI agent to understand global project context, perform cross-file edits, and execute or debug code in local or sandboxed environments.
| Tool | Pricing (2026-06) | Discounts / free credits | Best for | Link |
|---|---|---|---|---|
| **Cursor** | Hobby: Free (limited)<br>Pro: **$20/mo (includes $20 quota)**<br>Pro+: **$60 (includes $70 quota)**<br>Ultra: $200 (includes $400 quota)<br>Teams Standard: $40/seat (annual $32)<br>Premium: $120/seat | ✅ **New accounts receive 50% off the first month via any referral link** (Pro tier ≈ $10 first month; higher tiers eligible; **Alipay accepted** — remember to cancel auto-renew to avoid the standard $20 billing in the second month). Referral links have usage caps; search for active links.<br>· Annual plans offer ~20% savings.<br>· ⚠️ **GLM-5.2 free usage ended on 2026-06-26** (verified); calls now deduct from your subscription quota like other models. | Power users; multi-model flexibility and parallel agentic workflows; largest developer community | [cursor.com](https://cursor.com) |
| **Devin Desktop**（formerly Windsurf） | Free (light quota, unlimited Tab completions)<br>Pro: **$20/mo**<br>Max: $200/mo<br>Teams: $80/mo base + $40/full seat | ✅ **New users receive $10 AI credit** upon completing onboarding (connecting a Git repository); Pro and above tiers include free access to the SWE 1.6 model;<br>⏳ **Limited-Time Promo: GLM-5.2 and Kimi K2.7 models are free for Pro/Max/Teams users until 2026-07-05** | Delegating complex, long-running engineering tasks to Devin Cloud for automated PR creation | [devin.ai](https://devin.ai) |
| **Google Antigravity** | Free tier available<br>or via **Google AI Pro ($19.99/mo)**<br>AI Plus: $7.99/mo<br>AI Ultra: **$99.99** / **$200 (reduced from $250)** | Includes a generous free tier; top-tier Ultra pricing reduced from $249.99 to $200 in May 2026; compute pools reset every 5 hours (subject to a weekly cap); inline Tab completion is entirely free. | Multi-agent parallel workflows, built-in browser interaction, and full-stack web development | [antigravity.google](https://antigravity.google) |
| **Kiro**（AWS, spec-driven） | Free: 50 credits<br>Pro: **$20/mo (includes 1,000 credits)**<br>Pro+: $40 (includes 2,000 credits）<br>Pro Max: **$100 (includes 5,000 credits)**<br>Power: $200 (includes 10,000 credits) | Annual billing discounts available; overage billed at $0.04/credit.<br>· Model consumption multipliers: Auto 1.0×, Sonnet 1.3×, **Opus 4.8 2.2×**, Haiku 0.4×, GLM-5 0.5×, **Qwen3 Coder Next 0.05×** | Spec-driven development emphasizing highly maintainable production-grade code; AWS-centric engineering teams | [kiro.dev](https://kiro.dev) |
| **Trae**（ByteDance） | Free tier (5K autocompletions/month, 2 concurrent tasks)<br>Lite: **$3/mo**<br>Pro: **$10/mo** (includes 7-day trial)<br>Pro+: $30/mo<br>Ultra: $100/mo | ✅ **The regional version (trae.cn) is completely free** for individual developers (utilizing Doubao and DeepSeek models) | Budget-conscious developers seeking a Cursor-like user experience | [trae.ai](https://trae.ai) |
| **ZCode**（Zhipu） | Promotional free credits<br>or paid via GLM Coding Plan | ✅ **New users receive a 5-day free trial with 5M tokens/day** (GLM-5.2 3M and GLM-5-turbo 2M allocation); GLM Coding Plan subscribers receive a **1.5× quota multiplier** — active through **2026-06-30** | Developers seeking an official, deeply integrated IDE for the GLM model ecosystem | [zcode.z.ai](https://zcode.z.ai/cn/docs/welcome) |
| **CodeBuddy**（Tencent, international） | Free: 250 credits / 2 weeks<br>Pro: **$9.95/mo** (promo price, originally $19.90) or $119.40/year (1,000 credits/month)<br>Team: $40/seat/month | Provides 250 credits per 2 weeks; additional credit packs start at $9.95 per 1,000 credits | Global developers; standardized coding assistance within the Tencent Cloud ecosystem | [codebuddy.ai](https://www.codebuddy.ai/) |
| **CodeBuddy / WorkBuddy**（Tencent, regional） | Trial: **Free (500 credits/month baseline)** (rate-limited chat)<br>Standard: **¥99/month** (2,000 credits)<br>Flagship: ¥198/month<br>Exclusive: ¥316/month | ✅ Via [WorkBuddy invite link](https://www.workbuddy.cn/events/invite?inviteCode=fmfw2whmh): invitees get **2,000 signup credits** (valid 1 month) + onboarding up to **3,600 total**; **first paid purchase: both parties get 500 credits** (invitee once only, valid 6 months); trial tier also includes 500 credits/month baseline; now includes flagship **Kimi K3** (~**1.62×** credit multiplier);<br>⏳ **Limited-Time Promo: HY3 model free for all users (0 credits) through 2026-07-22** | Regional developers in the Tencent Cloud ecosystem; **new users wanting a free Kimi K3 trial should prefer this first — first-month credits are usually enough** | [workbuddy.cn](https://www.workbuddy.cn/events/invite?inviteCode=fmfw2whmh) / [codebuddy.cn](https://www.codebuddy.cn/pricing/) |
| **Qoder CN**（formerly Tongyi Lingma, Alibaba） | Community: **Free** (restricted quota, includes 2-week Pro trial and 300 credits)<br>Professional: ¥59/month (2,000 credits)<br>Enterprise: ¥99 or ¥199/seat/month | ⚠️ **Rebranded and repriced on 2026-05-20**; the free tier has been significantly restricted (limited completions and quotas), ending the previous unlimited-free era | Developers within the Alibaba Cloud ecosystem requiring flexible switching between regional models (Qwen, GLM, Kimi) | [cn.aliyun.com/product/lingma](https://cn.aliyun.com/product/lingma) |
| **Zed** | Editor core features free<br>Low-cost Pro subscription tier available | Core editor is open-source and free;<br>✅ **Verified students receive a free 1-year membership** including ~$10/month in AI model credits | Developers prioritizing extreme speed, local-first workflows, and strong privacy compliance | [zed.dev](https://zed.dev) |
⌨️ CLI & Terminal Agents
AI agents running natively in the terminal — capable of direct codebase editing, automated test execution, and Git workflow management.
| Tool | Pricing (2026-06) | Discounts / free | Best for | Link |
|---|---|---|---|---|
| **Freebuff** | **100% free** (ad-supported; CLI / Web / Chat / Desktop)<br>CLI: npm install -g freebuff | $0 forever — no API key or credit card; includes **DeepSeek V4 Flash/Pro** (incl. **Flash-0731**), Kimi, MiniMax, and more; [Web referral link](https://freebuff.com/web/?ref=ref-b3046190-1d77-44c1-915c-b71b4f36ae5e) | **Best $0 path to latest DeepSeek V4**; zero-budget CLI agent / web builder / chat | [freebuff.com](https://freebuff.com/) |
| **Codebuff** | Subscriptions: **$100/mo** (1× usage), **$200/mo** (2.5×), **$500/mo** (7×)<br>Pay-as-you-go: **500 free credits** on signup, then **$0.01/credit** | Signup credits (~few hours on a new project); cancel anytime | Premium terminal agent with deep codebase indexing and higher output quality; same company as Freebuff, but a paid product | [codebuff.com](https://www.codebuff.com) |
| **Claude Code** | Free tier does not include Claude Code<br>Pro: **$17/mo (annual)** or $20/mo (monthly)<br>Max: 5x $100 or 20x $200<br>Team: $20–$100/seat/month | Annual plans offer ~15% savings; usage limits are based on a rolling 5-hour window (quota doubled on 2026-05-06) shared with Claude Web Chat; Max tier features dual weekly limits | Advanced reasoning (Opus 4.8), large-scale architectural refactoring, and million-token context analysis | [claude.com/claude-code](https://www.claude.com/product/claude-code) |
| **Grok Build**（xAI） | Free tier of Grok chat (rate-limited)<br>**SuperGrok: $30/mo**<br>SuperGrok Heavy: higher tier | ✅ **7-day SuperGrok free trial** at [grok.com/plans](https://grok.com/plans) — includes **Grok 4.5** and Grok Build at paid limits; requires a credit card; **cancel before trial ends** to avoid $30/mo billing. Appears aimed at **never-subscribed** accounts; promo end date TBD | Terminal coding agent powered by Grok 4.5; evaluate frontier xAI models without an immediate subscription | [grok.com/plans](https://grok.com/plans) |
| **Kimi**（Moonshot / Kimi Code） | Adagio: **Free**<br>Andante: **¥49/mo**<br>Moderato: ¥99/mo<br>Allegretto: ¥199/mo<br>Allegro: ¥699/mo<br>(annual billing cheaper; shared credit pool for Kimi Code / Agent)<br>*Intl tiers also available (e.g. Moderato $19/mo)* | ⏳ **Limited-time free membership lottery** via [invite link](https://kimi-bot.com/activities/zh-cn/viral-referral/share?scenario=invite&from=share_poster&invitation_code=ZCDJXZ) — **3 / 7 / 15 / 30 / up to 365 days** (min. 3 days); flagship **Kimi K3** (**2026-07-17**, **TOP 3** on multiple benchmarks) live on Web / Code / API | Developers wanting membership access to Kimi Code / Agent, or evaluating the latest **Kimi K3** for coding & agentic work; **for a free K3 trial, prefer WorkBuddy first** (first-month credits usually suffice) | [kimi.com/membership/pricing](https://www.kimi.com/membership/pricing?from=kfc_membership_upgrade_price&track_id=fcbd4630-4439-498a-b934-d6cee41da4f1) |
| **OpenAI Codex** | **Included in ChatGPT Free** (Web, CLI, and desktop applications, subject to the lowest 5-hour rolling window limits)<br>Go: $8/mo<br>Plus: $20/mo<br>Pro: $100–$200/mo<br>Business/Enterprise: Custom pricing | $0 cost via ChatGPT Free with no credit card required; promotional limit boosts have concluded; supports integration with your personal **OpenAI API Key** (billed per-token, bypasses window limits, but excludes cloud sandbox features) | Deep integration with the OpenAI ecosystem; **desktop applications (macOS/Windows) are the recommended interface**; supports cloud sandboxing | [openai.com/codex](https://openai.com/codex) |
| **Gemini CLI / Antigravity CLI** | **Free tier** (provides high daily quotas) | Highly generous free-tier allocation | Light-to-moderate usage requiring massive context windows | [antigravity.google](https://antigravity.google) |
| **Ollama Cloud** | Free: $0<br>Pro: **$20/mo** (or $200/yr)<br>Max: $100/mo | Pro includes **50×** more cloud usage than Free and **3 concurrent cloud models** (Free: 1; Max: 10). Cloud usage is GPU-time based with **5-hour session** and **7-day weekly** reset windows. Local models remain unlimited. | Developers who want a single local-first workflow while running large cloud models (e.g. glm-5.2:cloud) without managing multiple vendor keys | [ollama.com/pricing](https://ollama.com/pricing) |
| **Aider** | **Free and open-source** (pay only for LLM API consumption, or $0 when running local models) | Completely free when paired with local open-source models | Git-native pair programming workflows; supports over 75+ commercial and open-source models | [aider.chat](https://aider.chat) |
| **Amp**（Sourcegraph） | **Pay-as-you-go, zero-markup** on raw API costs; $5 minimum initial deposit | Provides a free daily credit allocation (~$10/day, replenished hourly) — ⚠️ **this allocation has been paused or reduced for some users since May 2026**; fully ad-free since 2026 | Dynamic routing across multiple models (e.g., GPT-5.5, Claude Opus) with unconstrained context windows | [ampcode.com](https://ampcode.com) |
| **Cline** | Open-source version is free (BYOK)<br>**ClinePass: $9.99/mo** — now includes **Kimi K3** at official **~5× discounted access**, plus GLM-5.2, Kimi K2.7 Code, DeepSeek V4, Qwen3.7, MiniMax, MiMo, and more; CLI via npm i -g cline | ⏳ **ClinePass launch promo: $1.99/mo** when signing up via CLI (end date TBD — verify at [cline.bot](https://cline.bot) / [x.com/cline](https://x.com/cline)) | One of the lower-cost paid paths to **Kimi K3**; regional new users wanting a free trial should prefer WorkBuddy; available on Cline CLI & IDE | [cline.bot](https://cline.bot) |
| **OpenCode** | Open-source version is free (BYOK)<br>**Go Plan: $5 for the first month, then $10/mo** (~**$60/mo** equivalent usage: $12/5-hour, $30/week) | ✅ **Referral Promo: $5 credit for both parties** via [referral link](https://opencode.ai/go?ref=J9E8732NMP); includes **DeepSeek V4 Flash/Pro** — Flash gets full **$60** (official unit prices → ~**1/6** effective cost when fully used); Pro capped at **$15** | **Cheapest paid path to latest V4 Flash (0731) vs official metering**; unified key for GLM / Kimi / Qwen / DeepSeek; works with third-party agents | [opencode.ai](https://opencode.ai) |
| **Crush**（Charm） | **Free and open-source** (supports custom API keys) | $0 software cost when paired with your own API key for pay-as-you-go billing | Developers seeking an elegant Terminal User Interface (TUI) and flexible multi-model routing | [github.com/charmbracelet/crush](https://github.com/charmbracelet/crush) |
| **Goose**（Block） | **Free and open-source** (supports custom API keys or local execution) | $0 software cost when paired with your own API key for pay-as-you-go billing | Highly extensible local agents with native support for the Model Context Protocol (MCP) | [block.github.io/goose](https://block.github.io/goose) |
| **Qwen Code** | **Free and open-source** | $0 software licensing cost when running Qwen-series models | Developers relying on the Qwen model ecosystem who require maximum cost efficiency | [github.com/QwenLM/qwen-code](https://github.com/QwenLM/qwen-code) |
🧩 Assistants & IDE Extensions
Integrate seamlessly into your existing editor (such as VS Code or JetBrains) to provide inline code completions, interactive chat, and autonomous agent modes.
| Tool | Pricing (2026-06) | Discounts / free | Best for | Link |
|---|---|---|---|---|
| **GitHub Copilot** | Free (2,000 completions/month)<br>Pro: **$10/mo (includes 1,500 AI credits)**<br>Pro+: $39/mo (includes 7,000 credits)<br>Max: $100/mo (includes 20,000 credits)<br>Business: $19/user/month<br>Enterprise: $39/user/month | ⚠️ **New sign-ups for Pro/Pro+/Max paused since 2026-06-01** (Max is upgrade-only for existing users). 1 credit equals $0.01; **inline completions and Next Edit features remain free**. Students can access Copilot Pro for free (see Academic & Student Benefits). | GitHub-centric workflows; the standard choice for enterprise-grade compliance | [github.com/features/copilot](https://github.com/features/copilot) |
| **Continue.dev** | Core features **completely free** (supports BYOK or local Ollama deployments)<br>Team: ~$20/seat/month | Completely open-source and free to self-host | Teams prioritizing strict privacy compliance, cost control, and custom model integration | [continue.dev](https://www.continue.dev) |
| **Cline** | **Free and open-source** (supports BYOK or local models)<br>**ClinePass: $9.99/mo** — now includes **Kimi K3** at official **~5× discounted access**, plus GLM-5.2, Kimi K2.7 Code, DeepSeek V4, Qwen3.7, MiniMax, MiMo, and more; works on IDE extension & CLI | ⏳ **ClinePass launch promo: $1.99/mo** when signing up via npm i -g cline (end date TBD — verify at [cline.bot](https://cline.bot) / [x.com/cline](https://x.com/cline)); BYOK/local mode remains $0 | Developers seeking a highly autonomous VS Code agent; one of the lower fixed-cost paid paths to **Kimi K3** (regional new users wanting a free trial should prefer WorkBuddy) | [cline.bot](https://cline.bot) |
| **Roo Code** | **Free and open-source** (supports custom API keys) | $0 software cost when paired with your own API key for pay-as-you-go billing | Developers preferring the Cline architecture but requiring additional specialized agent modes | [roocode.com](https://roocode.com) |
| **Augment Code** | Free Community tier (restricted)<br>Trial: 30K credits (requires credit card)<br>Business: **$100/month flat rate** (covers up to 50 seats, includes $100 usage)<br>Enterprise: Custom pricing | Provides a 30K-credit trial; flat-rate Business pricing (no per-seat fees up to 50 seats) | Teams working with large, complex codebases requiring deep, global context | [augmentcode.com](https://www.augmentcode.com) |
| **Amazon Q Developer** | **Perpetual free tier** (includes 50 agentic requests/month and 1,000 lines of code transformation/month)<br>Pro: **$19/user/month** | Generous free tier covering both IDE and CLI with no credit card required | AWS-centric engineering teams; ideal for Java and .NET legacy codebase modernization | [aws.amazon.com/q/developer](https://aws.amazon.com/q/developer/) |
| **Supermaven** | Free tier (high-speed completions, supports large repositories)<br>Pro: **$10/month** (includes 1M context window, coding style adaptation, and $5 chat credits)<br>Team: $10/seat/month | Highly capable free tier; Pro plan offers a 30-day trial | Developers prioritizing near-zero latency and high-speed autocomplete performance | [supermaven.com](https://supermaven.com) |
| **Qodo**（formerly CodiumAI） | Free Developer tier (includes 30 organization-level PR reviews/month and 250 IDE/CLI credits)<br>Teams: $30/user/month (annual) or $38/user/month (monthly) | Highly practical free tier focusing on automated PR reviews and unit test generation | Technical teams focusing on automated test suite generation and PR workflow auditing | [qodo.ai](https://www.qodo.ai) |
🎯 aiskill88 AI 点评
A 级
2026-07-02
该项目提供了丰富的AI编码工具优惠信息，值得关注
📚 实用指南（长尾问题）
适合谁
- 使用 Cursor 编辑器、希望提升 AI 编程效率的开发者
- 需要让 Claude / Cursor 操作本地工具的 AI 工程师
- 构建多智能体协作系统的 Agent 开发者
最佳实践
- 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
- 本地部署优先选 GGUF 量化模型，节省显存并保持响应速度
- Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
- Cursor rules 控制在 80 行内，否则模型上下文成本会显著上升
常见错误
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
- 显存不足直接 OOM — 优先降低 context 或换更小的量化模型
- Python 依赖冲突：建议用 venv / uv 隔离环境
部署方案
- CLI：直接 npm install -g / pip install，命令行调用
- 本地部署：CPU 8GB 起，GPU 推荐 16GB+ 显存
- 云端托管：可放在 Vercel / Railway / Fly.io 等 PaaS 平台
⚡ 核心功能
- 开源免费，支持本地部署，数据完全自主可控
- 活跃的 GitHub 开源社区，持续迭代更新
- 提供详细文档和使用示例，新手友好
- 支持自定义配置，灵活适配不同使用环境
- 可作为基础组件集成进现有技术栈或进行二次开发
👥 适合谁
- 使用 Cursor 编辑器、希望提升 AI 编程效率的开发者
- 需要让 Claude / Cursor 操作本地工具的 AI 工程师
- 构建多智能体协作系统的 Agent 开发者
⭐ 最佳实践
- 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
- 本地部署优先选 GGUF 量化模型，节省显存并保持响应速度
- Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
- Cursor rules 控制在 80 行内，否则模型上下文成本会显著上升
⚠️ 常见错误
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
- 显存不足直接 OOM — 优先降低 context 或换更小的量化模型
- Python 依赖冲突：建议用 venv / uv 隔离环境
👥 适合人群
🎯 使用场景
- 本地部署运行，保护数据隐私，满足合规要求
- 自定义集成到现有系统，扩展技术栈能力
- 作为开源基础组件进行商业化二次开发
⚖️ 优点与不足
✅ 优点
- +MIT 协议，可免费商用
- +完全开源免费，无授权费用
- +本地部署，数据完全自主可控
- +开发者社区支持，遇问题可查可问
⚠️ 不足
- −安装和初始配置可能需要一定技术基础
- −功能完整性通常不如成熟商业产品
- −技术支持主要依赖开源社区，响应速度不稳定
⚠️ 使用须知
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
📄 License 说明
✅ MIT 协议 — 最宽松的开源协议之一，可自由商用、修改、分发，仅需保留版权声明。
🔗 相关工具推荐
yt-dlp 视频下载
功能强大的开源视频下载工具，支持YouTube、TikTok等数千个视频平台，可自动下载视频、字幕、封面和元数据。适合内
transformers AI技能包
Hugging Face开源的深度学习框架，提供预训练语言模型、视觉模型和多模态模型。集成BERT、GPT、Llama等
ComfyUI 节点式AI图像生成
强大的开源扩散模型可视化工具，提供图形界面、API和后端服务。采用节点图式设计，支持模块化工作流构建，适合AI绘图、图像
llama-cpp AI技能包
高效的大语言模型C/C++推理框架，支持在本地CPU/GPU上运行量化LLM模型，具有内存占用小、推理速度快的特点。适合
📰 相关 AI 新闻
🍿 AI 圈相关吃瓜
🧩 你可能还需要
基于当前 Skill 的能力图谱，自动补全的工具组合
❓ 常见问题 FAQ
ai-coding-deals 是一款Python开发的AI辅助工具。开源AI工具：Money-saving guide to AI coding tools: free tiers, discounts, referral credits &。⭐45 · Python 主要应用场景包括：寻找AI编码工具的优惠和折扣。
💡 AI Skill Hub 点评
总体来看，AI编码优惠 是一款质量优秀的AI工具，在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态，建议收藏备用，结合自身场景选择合适时机引入使用。
🌐 原始信息
| 原始名称 | ai-coding-deals |
| 原始描述 | 开源AI工具：Money-saving guide to AI coding tools: free tiers, discounts, referral credits &。⭐45 · Python |
| Topics | aiai-codingai-toolspython |
| GitHub | https://github.com/codertesla/ai-coding-deals |
| License | MIT |
| 语言 | Python |
🔗 原始来源
🐙 GitHub 仓库 https://github.com/codertesla/ai-coding-deals
收录时间：2026-07-02 · 更新时间：2026-07-03 · License：MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。