GLM vs Kimi vs Qwen
The three big lab-direct coding subscriptions. One publishes exact credits, one gates its best context behind the top tier, and one paused sign-ups entirely.
Updated August 28, 2026 · Independent comparison — GLM Coding Plan, Kimi Code, Qwen Coding Plan are separate products.
GLM Coding Plan
Z.ai's GLM-5.3 subscription for coding tools
The GLM Coding Plan is Z.ai's subscription for driving coding agents with GLM-5.3 and GLM-5.3-Flash. It sells in three tiers (Lite, Pro, Max) that differ only in credit allowance, not in model access, and it is explicitly restricted to coding tools — the key is not a general-purpose API key. Credits refresh on a 5-hour rolling window and a weekly window, and off-peak usage (outside Mon-Fri 14:00-18:00 UTC+8) burns credits at half rate. Z.ai relaunched the tier pricing on 2026-07-31, so list prices differ noticeably between its CNY and USD storefronts.
Kimi Code
Moonshot's Kimi K3 membership for coding
Kimi Code is the coding half of Moonshot's Kimi membership. Every paid tier includes it, but K3 access starts at the second tier and the full million-token K3 context only appears at the top. The membership meters work in 'agent credits' — roughly one typical task each — and Kimi Code additionally has its own separate 5-hour and weekly rate limits that Moonshot does not publish numerically. Concurrency is explicitly tiered (2 to 4 simultaneous agent tasks). Moonshot paused new consumer sign-ups on 2026-07-19 after K3 demand saturated its GPUs, reopening in batches.
Qwen Coding Plan
Alibaba's Qwen3.8 credit subscription
The Qwen Coding Plan — now sold as the QwenCloud Token Plan — is Alibaba's credit subscription for driving coding agents with Qwen3.8 Max and friends. It is unusual in two ways: the personal quota window is 7 days rather than the 5-hour window most rivals use, and the credit pool also covers non-text models (image, TTS, realtime audio) plus third-party weights like GLM-5.2 and DeepSeek V4. It issues a dedicated sk-sp- key that is not interchangeable with pay-as-you-go keys, and the plan documents a recommended concurrent-agent count per tier. Inference runs out of Singapore.
Bottom line
All three are lab-direct coding subscriptions, and they differ most in the shape of the meter. The GLM Coding Plan publishes exact credit quotas on a 5-hour and weekly window, supports the widest range of third-party agents, halves credits off-peak, and starts at $18. Kimi Code starts at $19, meters in 'agent credits' plus a separate unpublished 5-hour and weekly cap, gates Kimi K3's full million-token context to the top tier, and — importantly — Moonshot paused new consumer sign-ups on 2026-07-19 after K3 demand saturated its GPUs. The Qwen Coding Plan is the cheapest at $6-8, uses an unusual 7-day window that suits bursty work, bundles image, TTS and realtime audio into the same credit pool, and publishes recommended concurrency per tier — but individual plans currently cannot be cancelled.
Pricing
| Feature | GLM Coding Plan | Kimi Code | Qwen Coding Plan |
|---|---|---|---|
| Price/mo | Lite $18 · Pro $72-80 · Max $160-168/mo | $19 / $39 / $99 / $199/mo (annual $15 / $31 / $79 / $159) | Lite $8 ($6 promo) · Standard $25 ($18) · Pro $80 ($68)/mo |
| Intro promo | ~20-30% off on the CNY store and on quarterly/yearly terms | Annual billing saves up to $480/yr | Limited-time promo pricing on all three tiers |
| Billing shape | Credits on 5-hour + weekly windows | Agent credits + a separate Kimi Code rate limit | Credits on a 7-day window |
Models & Context
| Feature | GLM Coding Plan | Kimi Code | Qwen Coding Plan |
|---|---|---|---|
| Models included | GLM-5.3, GLM-5.3-Flash (older GLM auto-routes up) | Kimi K3, K2.7 Code, K2.6 | Qwen3.8 Max/Flash, Qwen3.7 Max/Plus, Qwen3.6 Flash, plus GLM-5.2, DeepSeek V4, image, TTS, realtime audio |
| Catalogue size | 2 models (one lab) | 3 models (one lab) | 14+ models across three labs and four modalities |
| Context window | Per model; no plan-level max published | K3 up to 1M — but only on the top tier | Per model; no plan-level max published |
Limits & Overage
| Feature | GLM Coding Plan | Kimi Code | Qwen Coding Plan |
|---|---|---|---|
| Limits (5h / weekly / monthly) | 2,000 / 12,000 / 28,000 credits per 5 hrs; 10,000 / 60,000 / 140,000 weekly | 60 / 150 / 360 / 720 agent credits; Kimi Code 5-hour + weekly caps unpublished | 2,500 / 10,000 / 40,000 credits per 7 days |
| What happens past the cap | Hard 429 until the window resets (errors 1308/1310/1316/1317) | In-flight tasks finish, new ones blocked; buy an Extra Usage pack or wait | Service pauses until the 7-day window ends (or buy a Credit Pack) |
| Parallel-agent friendly | Not published | 2 / 2 / 4 / 4 concurrent agent tasks by tier | Recommended 1-2 / 3-4 / 6-8 agents by tier |
Access & Friction
| Feature | GLM Coding Plan | Kimi Code | Qwen Coding Plan |
|---|---|---|---|
| Host agents & IDEs supported | Claude Code, Cline, Roo Code, Kilo Code, OpenCode, OpenClaw, Crush, Goose, Cursor, ZCode | Kimi Code, Kimi CLI, Claude Code, Roo Code, Kimi Claw | Claude Code, Cursor, Qwen Code, Qoder, OpenClaw, Codex |
| Gives you an API key | Coding Plan key — tool-restricted, not a general API key | Membership access is separate from the Open Platform API key | Dedicated sk-sp- key — coding tools only, one device |
| BYOK-able elsewhere | ✗ | ✗ | ✗ |
| Payment / region friction | Cards internationally; Alipay/WeChat on the CN store; no app-backend use | Cards internationally, Alipay/WeChat in China; new sign-ups paused 2026-07-19 | QwenCloud checkout; Singapore inference only; individual plans cannot be cancelled |
Verdict at a glance
| Feature | GLM Coding Plan | Kimi Code | Qwen Coding Plan |
|---|---|---|---|
| Best for | Cheapest official GLM-5.3 pass for supported agents | First-party K3 agent work if you can get a seat | Qwen3.8 Max plus multimodal on one credit pool |
The Verdict
Choose GLM Coding Plan if...
- ✓You want published, countable credit quotas
- ✓You want the widest third-party agent support in this group
- ✓Off-peak scheduling at half credits fits your working day
- ✓You want a clean Lite to Max ladder with a 14x spread
- ✓You want to be able to cancel normally
Choose Kimi Code if...
- ✓Kimi K3 specifically is the model you want
- ✓You want Moonshot's own tools — Kimi Code, Kimi CLI, Kimi Claw
- ✓You want published concurrency (2 or 4 simultaneous agent tasks)
- ✓You will pay the top tier for K3's full million-token context
- ✓You can wait for sign-ups to reopen in batches
Choose Qwen Coding Plan if...
- ✓Price is decisive — $6-8 against $18 and $19
- ✓Your work is bursty and a 7-day window absorbs a heavy day
- ✓You want image, TTS and realtime audio on the same credit pool
- ✓You want GLM-5.2 and DeepSeek V4 included alongside Qwen
- ✓You want published concurrency guidance before fanning out agents
Frequently Asked Questions
What is the difference between GLM Coding Plan, Kimi Code and Qwen Coding Plan?
The GLM Coding Plan is Z.ai's subscription for driving coding agents with GLM-5.3 and GLM-5.3-Flash. Kimi Code is the coding half of Moonshot's Kimi membership. The Qwen Coding Plan — now sold as the QwenCloud Token Plan — is Alibaba's credit subscription for driving coding agents with Qwen3.8 Max and friends.
How much do GLM Coding Plan, Kimi Code and Qwen Coding Plan cost?
GLM Coding Plan: Lite $18 · Pro $72-80 · Max $160-168/mo. Kimi Code: $19 / $39 / $99 / $199/mo (annual $15 / $31 / $79 / $159). Qwen Coding Plan: Lite $8 ($6 promo) · Standard $25 ($18) · Pro $80 ($68)/mo. See the pricing table above for full plan details.
Which of GLM Coding Plan, Kimi Code and Qwen Coding Plan should I pick?
Choose GLM Coding Plan if you want published, countable credit quotas. Choose Kimi Code if kimi K3 specifically is the model you want. Choose Qwen Coding Plan if price is decisive — $6-8 against $18 and $19.
1DevToolStop renting one agent's model pool
Every pass on this page locks a model pool to a particular agent or gateway, and each one meters you differently. 1DevTool runs Cline, OpenCode, Claude Code, Codex CLI and Gemini CLI side by side in persistent terminals — point each at whichever pass is cheapest this month, keep your own keys, and get an HTTP client, a 13-engine database client and an embedded browser in the same window. One-time $29, no subscription.