A complete comparison of China’s top 7 Coding Plans (Volcengine / Alibaba / Tencent / Baidu / Kimi / Zhipu / Minimax), including each platform’s models, pricing, quota, and features. This is basically the full picture of the current AI coding ecosystem in China.
A Coding Plan is an AI programming model subscription package offered by cloud vendors.
Core features:
- Fixed monthly fee
- Provides a large quota of AI requests
- Can be used in tools like Cursor / Claude Code / Cline / OpenClaw
In essence: Coding Plan = AI programming model subscription service.
1. Zhipu GLM Coding Plan
Platform: Zhipu AI
Supported models:
- GLM-4.7
- GLM-5
- GLM-5.1
- GLM-5-Turbo
- Latest Zhipu models
Features:
- Strong reasoning ability
- Agent ecosystem
- Supports 20+ programming tools including Claude Code
- Gradually rolling out latest flagship models and features
Link: https://www.bigmodel.cn/glm-coding?ic=HUPXXXNAM8

2. Volcengine Ark Coding Plan
Platform: Volcengine
Supported models:
- Doubao-Seed-Code
- DeepSeek-V3.2
- GLM-4.7
- Kimi-K2
Users can freely switch models or use Auto mode for automatic selection.
Features:
- Multi-model aggregation
- Auto model selection
- Compatible with Cursor / Claude Code / Cline
- Mainly focuses on ByteDance ecosystem models
Lite Plan:
- Per 5 hours: up to approx. 1,200 requests
- Per week: up to approx. 9,000 requests
- Per subscription month: up to approx. 18,000 requests
Pro Plan:
- 5x the quota of Lite Plan
- Per 5 hours: up to approx. 6,000 requests
- Per week: up to approx. 45,000 requests
- Per subscription month: up to approx. 90,000 requests
Subscription link: https://volcengine.com/L/0wCTjxOa9jA/

3. Alibaba Cloud Coding Plan
Platform: Alibaba Cloud
Supported models:
- Qwen3.5-Plus
- Qwen3-Coder
- GLM-5
- MiniMax-M2.5
- Kimi-K2.5
Features:
- Qwen coding models are very strong
- Large number of models
- Strong agent ecosystem
- Has usage restrictions (only allowed for AI coding tool calls)
Pro Plan:
- 6,000 requests per 5 hours
- 45,000 requests per week
- 90,000 requests per month
Subscription link: https://bailian.console.aliyun.com/

4. Tencent Cloud Coding Plan
Platform: Tencent Cloud
Supported models:
- Tencent HY 2.0
- GLM-5
- Kimi-K2.5
- MiniMax-M2.5
Pros:
- Tencent Hunyuan models
- Strong enterprise ecosystem
Cons:
- Programming model maturity slightly weaker than Qwen
Lite Plan: 40 yuan/month
- Per 5 hours: up to approx. 1,200 requests
- Per week: up to approx. 9,000 requests
- Per subscription month: up to approx. 18,000 requests
Pro Plan: 200 yuan/month
- Per 5 hours: up to approx. 6,000 requests
- Per week: up to approx. 45,000 requests
- Per subscription month: up to approx. 90,000 requests

5. Baidu Qianfan Coding Plan
Platform: Baidu
Supported models:
- GLM-5
- MiniMax-M2.5
- Kimi-K2.5
- Wenxin series models
Features:
- Qianfan ecosystem
- Many enterprise users
- Smaller developer community
Link: https://cloud.baidu.com/product/codingplan.html

6. Kimi Code
Platform: Moonshot AI
Core models:
- Kimi K2
- Kimi K2.5
Features:
- Long context
- Strong reasoning ability
Suitable for:
- Complex project analysis
- AI agents
Link: https://www.kimi.com/code

7. MiniMax Token Plan
I’ve also been using MiniMax for coding recently – the results are decent, and it serves as a good alternative to ChatGPT.
Supported models:
- MiniMax 2.7
- Speech-2.8-HD (speech model)
- MiniMax Hailuo 2.3 (video model)
- image-01 (image model)
- music-2.5+ (music model)
Features:
- M2.7 flagship for coding
- Extreme cost-effectiveness
- All-modal one-stop solution
Link: https://platform.minimaxi.com/subscribe/token-plan

Coding Plan Comparison
| Provider | Price | Recommendation | Features |
|---|---|---|---|
| Volcengine | 40 yuan/month | ByteDance series models, switch between multiple models | |
| Alibaba Cloud | 40 yuan/month | Qwen is the strongest, switch between multiple models | |
| Tencent Cloud | 40 yuan/month | Hunyuan ecosystem, switch between multiple models | |
| Baidu Cloud | 39.9 yuan/month | Enterprise ecosystem, switch between multiple models | |
| Kimi | 49 yuan/month | Long context | |
| Zhipu | 49 yuan/month | Strong reasoning, SOTA level | |
| MiniMax | 29 yuan/month | Extreme cost-effectiveness, SOTA level |