AI subscription plans compared
Prices and limits of AI chat and coding plans, checked by hand on each vendor's official page. No affiliate links.
Click a product to see what each plan includes.
| Product | Plan | Price | Main limits | ||
|---|---|---|---|---|---|
| ChatGPTOpenAIOfficial site ↗ | Go | — | CA$11/mo | — | More messages, uploads and image creation than Free; may include ads |
| Plus | $20/mo | CA$25/mo | — | Advanced reasoning models; expanded Codex and deep research | |
| Pro | from $100/mo | from CA$136/mo | — | 3 usage tiers (100 / 200 / 500); longer Codex sessions; monthly API credits | |
| Checked 2026-11-21 · Source ↗Checked 2026-11-21 · Source ↗See official site ↗ | |||||
| ClaudeAnthropicOfficial site ↗ | Pro | $20/mo | $20/mo* yearly $17/mo | — | More usage than Free; includes Claude Code |
| Max | — | $100/mo* | — | 5x Pro usage; $100/month API credits | |
| Ultra | — | $200/mo* | — | 10x Pro weekly usage, no 5-hour limit; $200/month API credits | |
| Checked 2026-11-21 · Source ↗Checked 2026-11-21 · Source ↗See official site ↗ | |||||
| CopilotGitHubOfficial site ↗ | Pro | $10/mo | US$10/mo | — | Unlimited completions; $15 monthly credits |
| Pro+ | $39/mo | US$39/mo | — | $70 monthly credits; premium models incl. Opus | |
| Max | $100/mo | US$100/mo | — | $200 monthly credits; priority access | |
| Checked 2026-11-21 · Source ↗Checked 2026-11-21 · Source ↗See official site ↗ | |||||
| Google AI (Gemini)GoogleOfficial site ↗ | Plus | $4.99/mo | CA$6.99/mo | — | 2x Gemini access vs. non-AI plans; 400 GB storage |
| Pro | $19.99/mo yearly $199.99/yr (≈$16.67/mo) | CA$26.99/mo | — | 4x Gemini access; 5 TB storage; YouTube Premium Lite where available | |
| Ultra 5x | $99.99/mo | CA$139.99/mo | — | 5x the Pro plan limits; 20 TB storage | |
| Ultra 20x | $199.99/mo | CA$279.99/mo | — | 20x the Pro plan limits; 30 TB storage; Project Genie | |
| Checked 2026-11-21 · Source ↗Checked 2026-11-21 · Source ↗See official site ↗ | |||||
| X Subscription (Grok + Cursor)xAIOfficial site ↗ | Plus | — | CA$9.99/mo | — | Base usage, one pool for Grok and the Grok models in Cursor; X Premium benefits; no Grok Bot. Other models in Cursor (e.g. Claude) are billed at API rates |
| Premium | — | CA$41.99/mo | — | 4x Plus usage; Grok Bot; latest Grok models; image and video generation | |
| Super | — | CA$139.99/mo | — | 15x Plus usage; priority at peak times | |
| Ultra | — | CA$279.99/mo | — | 50x Plus usage; most powerful models, earliest access | |
| See official site ↗Checked 2026-11-21 · excl. tax · Source ↗See official site ↗ | |||||
| PerplexityOfficial site ↗ | Pro | — | $17/mo* billed yearly | — | Deep research; choice of top models; expanded Computer access |
| Max | — | $167/mo* billed yearly | — | Highest limits; 10,000 monthly Computer credits; frontier models | |
| See official site ↗Checked 2026-11-21 · Source ↗See official site ↗ | |||||
| GLM Coding PlanZhipu / Z.ai ChinaOfficial site ↗ | Lite | — | $18/mo* yearly $12.6/mo | ¥118/mo quarterly ¥94.4/mo | 10,000 credits/week; works in Claude Code and 20+ tools |
| Pro | — | $80/mo* yearly $56/mo | ¥538/mo quarterly ¥430.4/mo | 6x Lite usage; curated MCP tools; faster | |
| Max | — | $168/mo* yearly $117.6/mo | ¥1078/mo quarterly ¥862.4/mo | 14x Lite usage; priority at peak times | |
| See official site ↗Checked 2026-11-21 · z.ai international site · Source ↗Checked 2026-11-21 · Source ↗ | |||||
| Kimi CodeMoonshot AI ChinaOfficial site ↗ | Plus | — | $19/mo* yearly $15/mo | — | Kimi Code; K3 model; multi-device login |
| Pro | — | $39/mo* yearly $31/mo | — | More limits and concurrency | |
| Max | — | $99/mo* yearly $79/mo | — | Generous credits for heavy development | |
| Ultra | — | $199/mo* yearly $159/mo | — | Highest credit allowance | |
| See official site ↗Checked 2026-11-21 · kimi.com international site · Source ↗See official site ↗ | |||||
| MiniMax M PlanMiniMax ChinaOfficial site ↗ | Go | — | $220/yr* yearly only · ≈$18.33/mo | ¥490/yr yearly only · ≈¥40.83/mo | Text models; MiniMax Code; API key for coding tools |
| Explore | — | $550/yr* yearly only · ≈$45.83/mo | ¥1190/yr yearly only · ≈¥99.17/mo | 3x Go usage; video generation | |
| Build | — | $1320/yr* yearly only · ≈$110/mo | ¥4690/yr yearly only · ≈¥390.83/mo | 12x Go usage (China site; 7.5x on international site); priority at peak | |
| See official site ↗Checked 2026-11-21 · minimax.io international site · Source ↗Checked 2026-11-21 · Source ↗ | |||||
| Bailian Token PlanAlibaba Cloud ChinaOfficial site ↗ | Lite | — | — | ¥39/mo | 20,000 credits/month; 1-2 concurrent agents |
| Essential | — | — | ¥79/mo | 45,000 credits/month; 2-3 concurrent agents | |
| Standard | — | — | ¥139/mo | 80,000 credits/month; 3-4 concurrent agents | |
| Pro | — | — | ¥312/mo | 200,000 credits/month; 6-8 concurrent agents | |
| See official site ↗See official site ↗Checked 2026-11-21 · list price; limited-time discounts shown as "from" · list price in the official documentation; lowered from 2026-11-09 for new subscription periods · Source ↗ | |||||
| Ark Coding PlanByteDance (Volcengine) ChinaOfficial site ↗ | Lite | — | — | — | Doubao, GLM, DeepSeek, Kimi models; Claude Code, Cursor |
| Pro | — | — | — | 5x Lite usage | |
| See official site ↗See official site ↗See official site ↗ | |||||
| Tencent Cloud Coding PlanTencent Cloud ChinaOfficial site ↗ | Lite | — | — | ¥40/mo | About 18,000 requests per subscription month (about 1,200 per 5 hours, 9,000 per week) |
| Pro | — | — | ¥200/mo | About 90,000 requests per subscription month (about 6,000 per 5 hours, 45,000 per week) | |
| See official site ↗See official site ↗Checked 2026-11-21 · List price per the official documentation (updated 2026-10-15); the new-customer promotion page has ended · Source ↗ | |||||
| Tencent Cloud Token PlanTencent Cloud ChinaOfficial site ↗ | Lite | — | — | ¥39/mo | 780 credits per subscription month; multiple Chinese models |
| Standard | — | — | ¥99/mo | 1,980 credits per subscription month; multiple Chinese models | |
| Pro | — | — | ¥299/mo | 5,980 credits per subscription month; multiple Chinese models | |
| Max | — | — | ¥599/mo | 11,980 credits per subscription month; multiple Chinese models | |
| See official site ↗See official site ↗Checked 2026-11-21 · Source ↗ | |||||
| Tencent Cloud Hy Token PlanTencent Cloud ChinaOfficial site ↗ | Lite | — | — | ¥28/mo | 560 credits per subscription month; Hunyuan models only |
| Standard | — | — | ¥78/mo | 1,560 credits per subscription month; Hunyuan models only | |
| Pro | — | — | ¥238/mo | 4,760 credits per subscription month; Hunyuan models only | |
| Max | — | — | ¥468/mo | 9,360 credits per subscription month; Hunyuan models only | |
| See official site ↗See official site ↗Checked 2026-11-21 · Hy Token Plan series; promotion page currently shows a limited-time 50% discount · Source ↗ | |||||
| Baidu Qianfan Token PlanBaidu AI Cloud ChinaOfficial site ↗ | Mini | — | — | ¥9.9/mo | 1,400 credits/month (credit billing) or 10M tokens/month (token billing) |
| Lite | — | — | ¥39/mo | 11,500 credits/month (credit billing) or 42M tokens/month (token billing) | |
| Standard | — | — | ¥100/mo | 22,000 credits/month; credit billing only | |
| Pro | — | — | ¥200/mo | 45,000 credits/month (credit billing) or 230M tokens/month (token billing) | |
| Max | — | — | ¥600/mo | 165,000 credits/month (credit billing) or 700M tokens/month (token billing) | |
| See official site ↗See official site ↗Checked 2026-11-21 · List price per the official documentation; ¥4.9 promotion until 2026-12-31 · List price per the official documentation; first-purchase promotion ¥19.9 (credit billing) until 2026-11-23 · List price per the official documentation; first-purchase promotion ¥49.9 (credit billing) until 2026-11-23 · List price per the official documentation; first-purchase promotion ¥99.9 (credit billing) until 2026-11-23 · List price per the official documentation; first-purchase promotion ¥299.9 (credit billing) until 2026-11-23 · Source ↗ | |||||
| Xiaomi MiMo Token PlanXiaomi ChinaOfficial site ↗ | Lite | — | US$6/mo yearly US$5.28/mo | ¥39/mo yearly ¥34/mo | 4.1B credits per month, shared across MiMo models |
| Standard | — | US$16/mo yearly US$14.08/mo | ¥99/mo yearly ¥87/mo | 11B credits per month, shared across MiMo models | |
| Pro | — | US$50/mo yearly US$44/mo | ¥329/mo yearly ¥289/mo | 38B credits per month, shared across MiMo models | |
| Max | — | US$100/mo yearly US$88/mo | ¥659/mo yearly ¥579/mo | 82B credits per month, shared across MiMo models | |
| See official site ↗Checked 2026-11-21 · MiMo official English pricing page (USD) · Source ↗Checked 2026-11-21 · Yearly plan ¥408/year; first personal purchase 12% off (monthly only, once per account) · Yearly plan ¥1,044/year; first personal purchase 12% off (monthly only, once per account) · Yearly plan ¥3,468/year; first personal purchase 12% off (monthly only, once per account) · Yearly plan ¥6,948/year; first personal purchase 12% off (monthly only, once per account) · Source ↗ | |||||
| Step PlanStepFun ChinaOfficial site ↗ | Plus | — | — | ¥99/mo yearly ¥78/mo · quarterly ¥89.67/mo | 1,600M credits per month (1M credits = ¥1); unused credits expire at the end of each period |
| Pro | — | — | ¥199/mo yearly ¥155/mo · quarterly ¥179.67/mo | 8,000M credits per month (1M credits = ¥1); unused credits expire at the end of each period | |
| Max | — | — | ¥399/mo yearly ¥355.5/mo · quarterly ¥376.33/mo | 20,000M credits per month (1M credits = ¥1); unused credits expire at the end of each period | |
| See official site ↗See official site ↗Checked 2026-11-21 · quarterly and yearly terms are paid up front; credits are issued monthly · Source ↗ | |||||
| Astron Coding PlaniFlytek ChinaOfficial site ↗ | Efficient | — | — | ¥199/mo quarterly ¥179.33/mo | About 6,000 requests per 5 hours, 45,000 per week, 90,000 per month |
| Express | — | — | ¥999/mo quarterly ¥999/mo | About 30,000 requests per month, with priority responses during peak hours | |
| See official site ↗See official site ↗Checked 2026-11-21 · official documentation price (updated 2026-07-22); the purchase page requires login · official documentation price (updated 2026-07-22); limited-time first-month offer ¥699 (¥2,697 for the first quarter), no end date stated; the purchase page requires login · Source ↗ | |||||
| Astron Token PlaniFlytek ChinaOfficial site ↗ | Standard member | — | — | ¥200/mo | 20,000 credits per member per month; 2M TPM |
| Advanced member | — | — | ¥600/mo | 60,000 credits per member per month; 3M TPM | |
| Premium member | — | — | ¥2000/mo | 200,000 credits per member per month; 5M TPM | |
| See official site ↗See official site ↗Checked 2026-11-21 · per member per month; for teams · Source ↗ | |||||
| Bailian Coding PlanAlibaba Cloud ChinaOfficial site ↗ | Pro | — | — | ¥200/mo | 6,000 requests/5 h; 45,000/week; 90,000/month |
| See official site ↗See official site ↗Checked 2026-11-21 · list price; new subscribers pay ¥39.90 for the first month (limited) · Source ↗ | |||||
| Ark Agent PlanByteDance (Volcengine) ChinaOfficial site ↗ | Small | — | — | ¥40/mo | 20,000 AFP/month; 7,000/week; 2,000/5 h |
| Medium | — | — | ¥200/mo | 100,000 AFP/month; 35,000/week; 10,000/5 h | |
| Large | — | — | ¥500/mo | 250,000 AFP/month; 87,500/week; 25,000/5 h | |
| Max | — | — | ¥1000/mo | 500,000 AFP/month; 175,000/week; 50,000/5 h | |
| See official site ↗See official site ↗Checked 2026-11-21 · List price. Limited-time offer on the activity page: first two months at a discount ("from ¥9.9"), limited availability. · Source ↗ | |||||
| TokenHub DeveloperChina Telecom (CTyun) ChinaOfficial site ↗ | Basic | — | — | ¥39.9/mo | 15 million tokens per month; GLM-5 |
| Professional | — | — | ¥159.9/mo | 70 million tokens per month; GLM-5 | |
| Flagship | — | — | ¥299.9/mo | 150 million tokens per month; GLM-5 | |
| See official site ↗See official site ↗Checked 2026-11-21 · Price per one-month token pack (activated manually, expires after one month, can be bought repeatedly); service stops when the quota runs out. · Source ↗ | |||||
| TokenHub PersonalChina Telecom (CTyun) ChinaOfficial site ↗ | Light | — | — | ¥9.9/mo | 10 million tokens per month; DeepSeek-V3.2 |
| Plus | — | — | ¥29.9/mo | 40 million tokens per month; DeepSeek-V3.2 | |
| Premium | — | — | ¥49.9/mo | 80 million tokens per month; DeepSeek-V3.2 | |
| See official site ↗See official site ↗Checked 2026-11-21 · Price per one-month token pack (activated manually, expires after one month, can be bought repeatedly); service stops when the quota runs out. · Source ↗ | |||||
* The page shows only “$” and does not name the currency.
≈ is the yearly price divided by 12, worked out by us; the vendor charges the full year up front.
Sorting puts plans in the region's own currency first; plans in other currencies follow, because we do not convert currencies.
Prices can differ by country, tax and promotion. The vendor's page is always the final word.
Frequently asked questions
Where do these prices come from?
From each vendor's own pricing page or help center. We open the page ourselves, copy the price by hand and record the date. Every product links to the page we used.
Why does the same plan cost different amounts in different regions?
Vendors set prices by country, often in local currency and sometimes with tax included. Pick your region at the top of the table; regions without a verified price show “See official site”.
What do “yearly” and “≈” mean?
“Yearly” is the monthly price when you pay for a whole year at once. When a vendor only gives the yearly total, we divide it by 12 and mark the result with ≈.
What does “$*” mean?
The vendor's page shows a dollar sign without saying which dollar. We mark it instead of guessing.
What is the difference between chat and coding plans?
Chat plans are for assistants you talk to in an app or browser. Coding plans are for coding tools such as editors and command-line agents, and are usually measured in requests or tokens.
How often is this updated?
We re-check every product about once a week, and on the same day when news of a price change comes in. Rows not checked for 14 days are marked as possibly out of date, and changes we find go on the Changes page.
Do you earn money from these links?
No. There are no affiliate links. If that ever changes, it will be stated on every page.
ChatGPT
OpenAI · Chat assistant · Coding
More messages, uploads and image creation than Free; may include ads
- Everything in Free
- More messages with tools
- More uploads
- More image creation
- More voice chats
- Longer memory
- May include ads
Advanced reasoning models; expanded Codex and deep research
- Everything in Go
- Advanced reasoning models with GPT-6
- Expanded messages and uploads
- More complex and accurate image creation
- Expanded deep research
- Expanded memory and context
- Projects, scheduled tasks and custom GPTs
- Expanded Codex usage
- Expanded ChatGPT Work on desktop, web and mobile
- Early access to new features
3 usage tiers (100 / 200 / 500); longer Codex sessions; monthly API credits
- Everything in Plus
- Choice of 3 usage tiers
- Pro reasoning powered by GPT-6 Astra
- Longer Codex and ChatGPT Work sessions
- Dot, an always-on agent
- Unlimited and faster image creation
- Maximum deep research
- Maximum memory and context
- Expanded projects, tasks and custom GPTs
- Monthly API credits (amount depends on tier)
Official site ↗ Full page: prices in every region and change log →
Claude
Anthropic · Chat assistant · Coding
More usage than Free; includes Claude Code
- Everything in Free
- More usage than Free, every 5 hours and every week
- Fable, paid with usage credits
- Claude Code and Claude Science
- Projects and more Claude models (incl. Opus)
- Research; hand off and schedule tasks
- Claude Data, plus Sheets and Whiteboards (beta)
- Claude in Chrome and Microsoft 365
5x Pro usage; $100/month API credits
- Everything in Pro
- 5x Pro usage every 5 hours and every week
- Fable included, up to half your weekly limit
- $100 a month in API credits for the Claude Platform
- Early access to advanced Claude features
- Priority access at high traffic times
10x Pro weekly usage, no 5-hour limit; $200/month API credits
- Everything in Max
- No 5-hour limit (verified accounts)
- 10x Pro usage every week
- $200 a month in API credits, up from $100 on Max
Official site ↗ Full page: prices in every region and change log →
Copilot
GitHub · Coding
Unlimited completions; $15 monthly credits
- Everything in Free
- Copilot in the cloud and code review
- Unlimited code completion and next edit suggestions
- 3rd party agents (Claude Code and Codex)
- Model selection
- $15 monthly total credits
$70 monthly credits; premium models incl. Opus
- Everything in Pro
- Premium models, including Opus
- 4x+ included usage than Pro
- $70 monthly total credits
- Copilot Fusion Standard, High, and Max
- Audit logs
$200 monthly credits; priority access
- Everything in Pro+
- Priority access to new models and features
- 2.9x+ included usage than Pro+
- $200 monthly total credits
Official site ↗ Full page: prices in every region and change log →
Google AI (Gemini)
Google · Chat assistant · Coding
2x Gemini access vs. non-AI plans; 400 GB storage
- Gemini 4 Hydrogen and Deep Research in Gemini
- Double the access to Gemini
- Gemini in Gmail to help you proofread
- More access to image, music and video generation
- More access to Gemini Notebook
- 200 Google Flow credits/mo
- 400 GB of cloud storage
- Family sharing with up to five others
4x Gemini access; 5 TB storage; YouTube Premium Lite where available
- Gemini 4 Neon; expanded Deep Research
- Four times more access to Gemini
- Gemini in Gmail, Docs and Sheets
- Expanded image, music and video generation, Google Flow
- Gemini 4 Neon in Google Antigravity and Google AI Studio
- 200 Colab compute units/mo; US$10 Google Cloud credits/mo
- YouTube Premium Lite (availability varies by country)
- 5 TB of cloud storage
5x the Pro plan limits; 20 TB storage
- Gemini 4 Carbon, Google's strongest model
- 5x higher usage limits in Gemini than AI Pro
- Higher access to image, music and video generation
- 10,000 Google Flow credits/mo
- Higher Gemini 4 Neon access in Antigravity and AI Studio
- US$40 Google Cloud credits/mo
- YouTube Premium Family (availability varies by country)
- 20 TB of cloud storage
20x the Pro plan limits; 30 TB storage; Project Genie
- Gemini 4 Carbon, Google's strongest model
- 20x usage limits in Gemini (vs Pro plan)
- Highest limits for generation, Flow and Antigravity
- 25,000 Google Flow credits/mo
- 2,000 Colab compute units/mo
- US$100 Google Cloud credits/mo
- Project Genie (interactive world model)
- 30 TB of cloud storage
Official site ↗ Full page: prices in every region and change log →
X Subscription (Grok + Cursor)
xAI · Chat assistant · Coding
Base usage, one pool for Grok and the Grok models in Cursor; X Premium benefits; no Grok Bot. Other models in Cursor (e.g. Claude) are billed at API rates
- One usage pool shared across all products
- Grok Chat, Voice, Imagine and Build
- Grok API
- Cursor (third-party models in Cursor not included)
- X: fewer timeline ads, Premium checkmark
- X: enhanced creator tools, advanced analytics
4x Plus usage; Grok Bot; latest Grok models; image and video generation
- Everything in Plus
- 4x the usage of Plus
- Grok Bot
- Latest Grok models, faster chat responses
- AI image and video generation
- Cursor cloud agents and automations
- Cursor code hosting and AI code reviews
- X: ad-free on most surfaces, preferred @handle, higher reply boost
15x Plus usage; priority at peak times
- Everything in Premium
- 15x the usage of Plus
- Priority model access at peak times
- Faster responses and generations
50x Plus usage; most powerful models, earliest access
- Everything in Super
- 50x the usage of Plus
- Most powerful chat models
- Earliest access to all products and models
- Fastest responses and generations
Official site ↗ Full page: prices in every region and change log →
Perplexity
Chat assistant
Deep research; choice of top models; expanded Computer access
- Everything in Free
- Expanded Computer access
- Deep research
- Access to top AI models; select between models
- Create polished documents & apps
- More usage limits and memory
Highest limits; 10,000 monthly Computer credits; frontier models
- Everything in Pro
- Maximum Computer usage
- 10,000 monthly credits
- Expert level research
- Frontier AI models
- Highest usage limits and memory
- Priority access to new features
Official site ↗ Full page: prices in every region and change log →
GLM Coding Plan
Zhipu / Z.ai · Coding China
10,000 credits/week; works in Claude Code and 20+ tools
- 10,000 credits / week
- Rolling access to latest flagship models and features
- Supports 20+ coding tools incl. ZCode, Claude Code
- Usage limits every 5 hours and every week
6x Lite usage; curated MCP tools; faster
- 6× Lite usage
- Everything in Lite
- Priority access to latest flagship models and features
- Includes a curated selection of MCP tools
- Faster generation speed
14x Lite usage; priority at peak times
- 14× Lite usage
- Everything in Pro
- First access to latest flagship models and features
- Dedicated resources prioritized at peak times
Official site ↗ Full page: prices in every region and change log →
Kimi Code
Moonshot AI · Coding China
Kimi Code; K3 model; multi-device login
- Regularly refreshed usage quotas
- Multi-device login, share plan across devices
- Work efficiently on multiple projects
- K3 available
- Also includes other Kimi membership benefits
More limits and concurrency
- Ample limits
- Increased concurrency caps
- K3 available
- Also includes other Kimi membership benefits
Generous credits for heavy development
- Generous credits
- For everyday work and intensive development
- K3 available
- Also includes other Kimi membership benefits
Highest credit allowance
- Highest credit allowance
- For complex projects and large codebases
- K3 available
- Also includes other Kimi membership benefits
Official site ↗ Full page: prices in every region and change log →
MiniMax M Plan
MiniMax · Coding China
Text models; MiniMax Code; API key for coding tools
- More usage than Free
- Text models, e.g. M3.1 Flash
- Image and speech generation
- Web search and multimodal understanding
- MiniMax Code on web, desktop, TUI, iOS, Android
- API key for AI agents and coding tools
- 15 scheduled tasks, 50 site deployments, 50 projects
3x Go usage; video generation
- 3× Go usage
- Text models, e.g. M3.1 Flash
- Video, image and speech generation
- Web search and multimodal understanding
- MiniMax Code on web, desktop, TUI, iOS, Android
- API key for AI agents and coding tools
- 20 scheduled tasks, 100 site deployments, 100 projects
12x Go usage (China site; 7.5x on international site); priority at peak
- 12× Go usage (7.5× on the international site)
- Text models, e.g. M3.1 Flash
- Video, image and speech generation
- Web search and multimodal understanding
- MiniMax Code on web, desktop, TUI, iOS, Android
- API key for AI agents and coding tools
- Priority at peak hours, early access to features
- 50 scheduled tasks, 200 site deployments, 200 projects
Official site ↗ Full page: prices in every region and change log →
Bailian Token Plan
Alibaba Cloud · Coding China
20,000 credits/month; 1-2 concurrent agents
- 20,000 credits / month
- 1–2 concurrent agents
- 20+ flagship models, incl. Qwen4-Max
- One key for Qwen Code, Qoder CN, OpenClaw, etc.
45,000 credits/month; 2-3 concurrent agents
- 45,000 credits / month
- 2–3 concurrent agents
- 20+ flagship models, incl. Qwen4-Max
- One key for Qwen Code, Qoder CN, OpenClaw, etc.
80,000 credits/month; 3-4 concurrent agents
- 80,000 credits / month
- 3–4 concurrent agents
- 12 Harness tools
- 20+ flagship models, incl. Qwen4-Max
- One key for Qwen Code, Qoder CN, OpenClaw, etc.
200,000 credits/month; 6-8 concurrent agents
- 200,000 credits / month
- 6–8 concurrent agents
- Includes Harness benefits
- 20+ flagship models, incl. Qwen4-Max
- One key for Qwen Code, Qoder CN, OpenClaw, etc.
Official site ↗ Full page: prices in every region and change log →
Ark Coding Plan
ByteDance (Volcengine) · Coding China
Doubao, GLM, DeepSeek, Kimi models; Claude Code, Cursor
- Models: Doubao, GLM, DeepSeek, Kimi, etc.
- Switch models freely or use Auto mode
- Tools: Claude Code, Cursor and other major tools
- Quota resets next cycle; no balance charged
5x Lite usage
- 5× Lite usage
- Everything in Lite
Official site ↗ Full page: prices in every region and change log →
Tencent Cloud Coding Plan
Tencent Cloud · Coding China
About 18,000 requests per subscription month (about 1,200 per 5 hours, 9,000 per week)
- Up to about 1,200 requests per 5 hours
- Up to about 9,000 requests per week
- Up to about 18,000 requests per subscription month
- Works in OpenClaw, CodeBuddy Code, Claude Code, Cline, Cursor, Codex and other tools
- Model selection: Auto (automatic matching)
- Dedicated API key; interactive use in coding tools only
- No multimodal support
About 90,000 requests per subscription month (about 6,000 per 5 hours, 45,000 per week)
- Up to about 6,000 requests per 5 hours
- Up to about 45,000 requests per week
- Up to about 90,000 requests per subscription month
- Works in OpenClaw, CodeBuddy Code, Claude Code, Cline, Cursor, Codex and other tools
- Model selection: Auto (automatic matching)
- Dedicated API key; interactive use in coding tools only
- No multimodal support
Official site ↗ Full page: prices in every region and change log →
Tencent Cloud Token Plan
Tencent Cloud · Coding China
780 credits per subscription month; multiple Chinese models
- 780 credits per subscription month
- Estimated about 70 conversational rounds in basic OpenClaw use
- Switch freely among MiniMax, GLM, DeepSeek, Tencent Hunyuan and other models
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- Interactive use in AI tools only; no refunds, upgrade only
1,980 credits per subscription month; multiple Chinese models
- 1,980 credits per subscription month
- Estimated about 200 conversational rounds in basic OpenClaw use
- Switch freely among MiniMax, GLM, DeepSeek, Tencent Hunyuan and other models
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- Interactive use in AI tools only; no refunds, upgrade only
5,980 credits per subscription month; multiple Chinese models
- 5,980 credits per subscription month
- Token quota 3x that of Standard
- Switch freely among MiniMax, GLM, DeepSeek, Tencent Hunyuan and other models
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- Interactive use in AI tools only; no refunds, upgrade only
11,980 credits per subscription month; multiple Chinese models
- 11,980 credits per subscription month
- Largest quota, for heavy AI development
- Switch freely among MiniMax, GLM, DeepSeek, Tencent Hunyuan and other models
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- Interactive use in AI tools only; no refunds, upgrade only
Official site ↗ Full page: prices in every region and change log →
Tencent Cloud Hy Token Plan
Tencent Cloud · Coding China
560 credits per subscription month; Hunyuan models only
- 560 credits per subscription month
- Estimated about 70 conversational rounds in basic OpenClaw use
- Tencent Hunyuan models only: Hy4 flash, Hy4, Hy3
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- No image or video input; no refunds, upgrade only
1,560 credits per subscription month; Hunyuan models only
- 1,560 credits per subscription month
- Estimated about 200 conversational rounds in basic OpenClaw use
- Tencent Hunyuan models only: Hy4 flash, Hy4, Hy3
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- No image or video input; no refunds, upgrade only
4,760 credits per subscription month; Hunyuan models only
- 4,760 credits per subscription month
- Token quota 3x that of Standard
- Tencent Hunyuan models only: Hy4 flash, Hy4, Hy3
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- No image or video input; no refunds, upgrade only
9,360 credits per subscription month; Hunyuan models only
- 9,360 credits per subscription month
- Largest quota, for heavy AI development
- Tencent Hunyuan models only: Hy4 flash, Hy4, Hy3
- Works in OpenClaw, Claude Code, OpenCode, Cline, Cursor, Kilo Code, Codex CLI
- Usage measured in credits (since 2026-08-31)
- No image or video input; no refunds, upgrade only
Official site ↗ Full page: prices in every region and change log →
Baidu Qianfan Token Plan
Baidu AI Cloud · Coding China
1,400 credits/month (credit billing) or 10M tokens/month (token billing)
- 1,400 credits per month (credit billing)
- 10M tokens per month (token billing)
- Credit-based or token-based billing (see note on Standard)
- Supports GLM, DeepSeek, ERNIE and other models
- Works in OpenClaw, Claude Code, OpenCode, Cursor, Cline, Codex and other tools
- Interactive use in AI tools only; no refunds or downgrades
- 5-hour and 7-day credit limits currently lifted for a limited time
11,500 credits/month (credit billing) or 42M tokens/month (token billing)
- 11,500 credits per month (credit billing)
- 42M tokens per month (token billing)
- Credit-based or token-based billing (see note on Standard)
- Supports GLM, DeepSeek, ERNIE and other models
- Works in OpenClaw, Claude Code, OpenCode, Cursor, Cline, Codex and other tools
- Interactive use in AI tools only; no refunds or downgrades
- 5-hour and 7-day credit limits currently lifted for a limited time
22,000 credits/month; credit billing only
- 22,000 credits per month
- Credit billing only
- Supports GLM, DeepSeek, ERNIE and other models
- Works in OpenClaw, Claude Code, OpenCode, Cursor, Cline, Codex and other tools
- Interactive use in AI tools only; no refunds or downgrades
- 5-hour and 7-day credit limits currently lifted for a limited time
45,000 credits/month (credit billing) or 230M tokens/month (token billing)
- 45,000 credits per month (credit billing)
- 230M tokens per month (token billing)
- Credit-based or token-based billing (see note on Standard)
- Supports GLM, DeepSeek, ERNIE and other models
- Works in OpenClaw, Claude Code, OpenCode, Cursor, Cline, Codex and other tools
- Interactive use in AI tools only; no refunds or downgrades
- 5-hour and 7-day credit limits currently lifted for a limited time
165,000 credits/month (credit billing) or 700M tokens/month (token billing)
- 165,000 credits per month (credit billing)
- 700M tokens per month (token billing)
- Credit-based or token-based billing (see note on Standard)
- Supports GLM, DeepSeek, ERNIE and other models
- Works in OpenClaw, Claude Code, OpenCode, Cursor, Cline, Codex and other tools
- Interactive use in AI tools only; no refunds or downgrades
- 5-hour and 7-day credit limits currently lifted for a limited time
Official site ↗ Full page: prices in every region and change log →
Xiaomi MiMo Token Plan
Xiaomi · Coding China
4.1B credits per month, shared across MiMo models
- 4.1B credits per month
- About 200 medium-to-complex tasks per month with mimo-v2.6-flash
- All plans include mimo-v2.6-pro, mimo-v2.6-flash, ASR and TTS models
- Works in OpenCode, OpenClaw, Claude Code and other tools
- TTS models free for a limited time
- Off-peak (00:00-08:00 Beijing time) consumption factor 0.8x
11B credits per month, shared across MiMo models
- 11B credits per month
- About 1,600 medium-to-complex tasks per month with mimo-v2.6-flash
- All plans include mimo-v2.6-pro, mimo-v2.6-flash, ASR and TTS models
- Works in OpenCode, OpenClaw, Claude Code and other tools
- TTS models free for a limited time
- Off-peak (00:00-08:00 Beijing time) consumption factor 0.8x
38B credits per month, shared across MiMo models
- 38B credits per month
- About 5,600 medium-to-complex tasks per month with mimo-v2.6-flash
- Also supports mimo-v2.6-pro-ultraspeed
- All plans include mimo-v2.6-pro, mimo-v2.6-flash, ASR and TTS models
- Works in OpenCode, OpenClaw, Claude Code and other tools
- TTS models free for a limited time
- Off-peak (00:00-08:00 Beijing time) consumption factor 0.8x
82B credits per month, shared across MiMo models
- 82B credits per month
- About 12,800 medium-to-complex tasks per month with mimo-v2.6-flash
- Also supports mimo-v2.6-pro-ultraspeed
- All plans include mimo-v2.6-pro, mimo-v2.6-flash, ASR and TTS models
- Works in OpenCode, OpenClaw, Claude Code and other tools
- TTS models free for a limited time
- Off-peak (00:00-08:00 Beijing time) consumption factor 0.8x
Official site ↗ Full page: prices in every region and change log →
Step Plan
StepFun · Coding China
1,600M credits per month (1M credits = ¥1); unused credits expire at the end of each period
- All flagship models
- Smart routing
- MCP tools
- Multi-device login
- Priority API rate limits
- Priority technical support
- Extra Studio creation quota equal to 40% of the plan quota
- Step 5 Preview uses 2x credits at weekday peak hours (Beijing time)
8,000M credits per month (1M credits = ¥1); unused credits expire at the end of each period
- All flagship models
- Smart routing
- MCP tools
- Multi-device login
- Priority API rate limits
- Priority technical support
- Extra Studio creation quota equal to 40% of the plan quota
- Step 5 Preview uses 2x credits at weekday peak hours (Beijing time)
20,000M credits per month (1M credits = ¥1); unused credits expire at the end of each period
- All flagship models
- Smart routing
- MCP tools
- Multi-device login
- Priority API rate limits
- Priority technical support
- Extra Studio creation quota equal to 40% of the plan quota
- Step 5 Preview uses 2x credits at weekday peak hours (Beijing time)
Official site ↗ Full page: prices in every region and change log →
Astron Coding Plan
iFlytek · Coding China
About 6,000 requests per 5 hours, 45,000 per week, 90,000 per month
- Works in Claude Code, Cursor, OpenCode, OpenClaw and Codex
- OpenAI- and Anthropic-compatible endpoints
- Spark X2 Agent, GLM-5.2, DeepSeek-V4-Pro, Kimi-K2.6 and other models
- Each model consumes requests at its own rate (factor 1 to 5)
- Interactive coding tools only; scripts and app backends are not permitted
- Upgrade to the Express tier only; no refunds
About 30,000 requests per month, with priority responses during peak hours
- Requests use the Express quota to reduce queuing and timeouts at peak hours
- Works in Claude Code, Cursor, OpenCode, OpenClaw and Codex
- OpenAI- and Anthropic-compatible endpoints
- Spark X2 Agent, GLM-5.2, DeepSeek-V4-Pro, Kimi-K2.6 and other models
- Each model consumes requests at its own rate (factor 1 to 5)
- Interactive coding tools only; scripts and app backends are not permitted
Official site ↗ Full page: prices in every region and change log →
Astron Token Plan
iFlytek · Coding China
20,000 credits per member per month; 2M TPM
- Unified credit metering across multiple models
- OpenAI- and Anthropic-compatible endpoints; works in OpenClaw, Cursor and Claude Code
- Off-peak usage consumes credits at 0.8x (weekdays outside 08:00–22:00, all weekend)
- Includes a one-month AstronClaw Basic membership (valued at ¥168)
- Includes iFlytek Skills such as speech synthesis, OCR and translation
- No upgrades, downgrades or refunds; multiple seats may be stacked
60,000 credits per member per month; 3M TPM
- Unified credit metering across multiple models
- OpenAI- and Anthropic-compatible endpoints; works in OpenClaw, Cursor and Claude Code
- Off-peak usage consumes credits at 0.8x (weekdays outside 08:00–22:00, all weekend)
- Includes a one-month AstronClaw Basic membership (valued at ¥168)
- Includes iFlytek Skills such as speech synthesis, OCR and translation
- No upgrades, downgrades or refunds; multiple seats may be stacked
200,000 credits per member per month; 5M TPM
- Unified credit metering across multiple models
- OpenAI- and Anthropic-compatible endpoints; works in OpenClaw, Cursor and Claude Code
- Off-peak usage consumes credits at 0.8x (weekdays outside 08:00–22:00, all weekend)
- Includes a one-month AstronClaw Basic membership (valued at ¥168)
- Includes iFlytek Skills such as speech synthesis, OCR and translation
- No upgrades, downgrades or refunds; multiple seats may be stacked
Official site ↗ Full page: prices in every region and change log →
Bailian Coding Plan
Alibaba Cloud · Coding China
6,000 requests/5 h; 45,000/week; 90,000/month
- 6,000 requests per 5 hours
- 45,000 requests per week; 90,000 per month
- Models: qwen3.7-plus, qwen3.6-plus, kimi-k2.5, glm-5, MiniMax-M2.5
- For coding tools only (Claude Code, Qoder, OpenClaw, etc.); API use is prohibited
- Requires the plan-specific API key (sk-sp-)
- Inputs and outputs may be used for service improvement
Official site ↗ Full page: prices in every region and change log →
Ark Agent Plan
ByteDance (Volcengine) · Coding China
20,000 AFP/month; 7,000/week; 2,000/5 h
- 20,000 AFP per month
- Weekly 7,000 AFP; 5-hour 2,000 AFP
- Popular coding models
- Seedream-5.0 series and embedding models
- Web search Harness included
- No video generation
100,000 AFP/month; 35,000/week; 10,000/5 h
- 5× Small usage
- Everything in Small
- Multimodal and web search Harness
- No video generation
250,000 AFP/month; 87,500/week; 25,000/5 h
- 12.5× Small usage
- Everything in Medium
- Supports Seedance-2.5 (video generation)
- Advanced multimodal and web search Harness
500,000 AFP/month; 175,000/week; 50,000/5 h
- 25× Small usage
- Everything in Large
- Supports Seedance-2.5 (video generation)
- Top-tier multimodal and web search Harness
Official site ↗ Full page: prices in every region and change log →
TokenHub Developer
China Telecom (CTyun) · Coding China
15 million tokens per month; GLM-5
- 15 million tokens
- Model: GLM-5, 200K context
- For developers and small businesses
- Works with OpenClaw, OpenCode, Cursor, Cline, Trae, Codex, etc.
- Peak hours 09:00–18:00; off-peak use recommended
70 million tokens per month; GLM-5
- 70 million tokens
- Model: GLM-5, 200K context
- For developers and small businesses
- Works with OpenClaw, OpenCode, Cursor, Cline, Trae, Codex, etc.
- Peak hours 09:00–18:00; off-peak use recommended
150 million tokens per month; GLM-5
- 150 million tokens
- Model: GLM-5, 200K context
- For developers and small businesses
- Works with OpenClaw, OpenCode, Cursor, Cline, Trae, Codex, etc.
- Peak hours 09:00–18:00; off-peak use recommended
Official site ↗ Full page: prices in every region and change log →
TokenHub Personal
China Telecom (CTyun) · Coding China
10 million tokens per month; DeepSeek-V3.2
- 10 million tokens
- Model: DeepSeek-V3.2
- For individuals and households: office assistance, study, writing
- Input and output tokens deducted at a single rate
- Real-time usage details
40 million tokens per month; DeepSeek-V3.2
- 40 million tokens
- Model: DeepSeek-V3.2
- For individuals and households: office assistance, study, writing
- Input and output tokens deducted at a single rate
- Real-time usage details
80 million tokens per month; DeepSeek-V3.2
- 80 million tokens
- Model: DeepSeek-V3.2
- For individuals and households: office assistance, study, writing
- Input and output tokens deducted at a single rate
- Real-time usage details
Official site ↗ Full page: prices in every region and change log →