Need a China coding plan (GLM / MiniMax / Kimi / Volcengine / Xiaomi)? Email me at [email protected]
Alibaba Cloud · Qwen
The request-based Coding Plan is now a Credits-based Token Plan: Qwen3.8-Max (2.4T, 1M context, 0902 snapshot) is generally available, joined by qwen3.8-flash, deepseek-v4-pro-0813, Kimi-K3 and the image / audio families in one Credits pool. Personal plans start at ¥39/mo with monthly Credits and 12 Harness benefits; night calls run at a limited-time 40% rate from 22:00.
Qwen3.8-Max on the 0902 snapshot, plus qwen3.8-flash, deepseek-v4-pro-0813, Kimi-K3, GLM-5.2 and the Qwen vision / image / audio families.
Alibaba’s 2.4T next-generation flagship on the 0902 snapshot release, generally available (the preview was retired; old model IDs auto-route). 1M context; officially described as capable of autonomous coding for days to deliver complete projects.
The new low-cost high-speed tier of the Qwen3.8 family.
DeepSeek-V4-Pro 0813 version snapshot, newly added.
Qwen3.7 reasoning and general-purpose flagships.
Image generation (qwen-image-3.0-pro) plus real-time audio models (Qwen3-TTS-Flash, Cosyvoice-V3-Flash, Fun-ASR) and wan2.7 image / happyhorse-1.1 video, all sharing the same Credits.
Third-party flagships Kimi-K3 and GLM-5.2 plus the Qwen3-VL-Plus vision model are part of the pool, sharing the same Credits.
Early-bird prices (Lite 35% / Standard 23% / Pro 17% off) with monthly Credits (11,500 / 45,000 / 180,000). 12 Harness benefits, an 88% over-limit rate and team seats are also available.
Lite
Entry ¥39/mo11,500 Credits/mo · elastic overage
Light users and first-timers
Standard
Recommended45,000 Credits/mo · 12 Harness benefits
Daily development workhorse
Pro
Heavy duty180,000 Credits/mo · overage 12% off
Heavy and multi-Agent workloads
| Tier | Monthly | Credits/mo | Agents | Notes |
|---|---|---|---|---|
| Lite | ¥39 | 11,500 | 1-2 | Elastic overage |
| Standard | ¥139 | 45,000 | 3-4 | 12 Harness benefits |
| Pro | ¥499 | 180,000 | 6-8 | Overage 12% off |
| Team standard seat | ¥150+ | 25,000 | — | List ¥198 |
| Team advanced seat | ¥550+ | 100,000 | — | List ¥698 |
| Team premium seat | ¥1,398 | 250,000 | — | — |
Personal plans now grant Credits monthly — Lite 11,500, Standard 45,000, Pro 180,000 — replacing the old 7-day window.
From 22:00, Qwen3.8-Max and Flash consume Credits at 40% of the normal rate (a limited-time 60% discount) — schedule heavy jobs accordingly.
In-plan calls are not used for model training.
Personal plans add 12 Harness benefits that bundle more models and agent capabilities into one subscription.
Over-limit calls are charged at 88% (12% off). Top-up packs: self-developed MCP ¥10, knowledge base (RAG) ¥20, AutoModel ¥20, personal usage pack ¥100, team usage pack ¥5,000.
Token Plan × Qoder CN: ¥98 (Token Plan Lite ¥39 + Qoder personal Pro ¥59) and ¥308 (Token Plan Standard ¥139 + Qoder personal Pro+ ¥169).
Alibaba Cloud upgraded the request-count based Coding Plan into a Credits-based Token Plan: models expanded from Qwen3.5 to 18+ flagship multimodal models led by Qwen3.8-Max (0902 snapshot) — plus qwen3.8-flash, deepseek-v4-pro-0813, Kimi-K3, GLM-5.2 and vision / image / audio families — with monthly Credits, 12 Harness benefits and personal/team versions.
Lite ¥39/mo (11,500 Credits/mo) for light use, Standard ¥139/mo (45,000 Credits/mo, ~3.9x Lite) as the daily workhorse with the 12 Harness benefits, Pro ¥499/mo (180,000 Credits/mo, ~15.7x Lite) for heavy multi-Agent workloads with overage billed at 12% off.
Each model has its own Credits rate, accumulated monthly. From 22:00, Qwen3.8-Max and Flash consume Credits at a limited-time 40% rate (60% off), and over-limit calls are billed at 88%.
Yes — team seats: Standard ¥150/seat/mo (list ¥198, 25,000 Credits/mo), Advanced ¥550 (list ¥698, 100,000) and Premium ¥1,398 (250,000), with a 29-model pool for centralized company billing; the team tiers were cut in price.
Personal plans add 12 Harness benefits that bundle more models and agent capabilities into one subscription, and over-limit calls are charged at 88% (12% off). Top-up packs are also available: self-developed MCP ¥10, knowledge base (RAG) ¥20, AutoModel ¥20, personal usage pack ¥100 and team usage pack ¥5,000.
Yes. Token Plan × Qoder CN bundles: ¥98 (Token Plan Lite ¥39 + Qoder personal Pro ¥59) and ¥308 (Token Plan Standard ¥139 + Qoder personal Pro+ ¥169).
It no longer applies. The limited-restock model of the old Coding Plan Pro was retired with the upgrade; Token Plan can be subscribed directly.
Activated plans are generally non-refundable. Confirm the rules before purchase.