ZCode
ActiveAgent-first development environment with GLM-5.3 and multi-agent collaboration
ZCode pairs an agent-first design with GLM-5.3 and GLM-5.3-Flash, on GLM Coding plans that start at $18 a month per Z.ai's docs. GLM-5.3 weights are published on Hugging Face, though under a bespoke licence to read before commercial use. Against that, Linux support is beta and reviewers have flagged credential handling for remote and messaging-triggered tasks. Conditional until security and trust catch up with the product.
The agent loop is the model: reads a step, acts, reads the next, holding full context across multi-step plans, execution, and verification.
Is it right for you?
Good for
- Long-horizon autonomous coding tasks where the agent plans, executes, and verifies across multiple iterations
- Deep GLM-5.3 integration, with GLM-5.3-Flash built in for multimodal tasks and no manual endpoint configuration
- Budget-conscious teams: GLM Coding Lite lists at $18/mo and is currently shown discounted to $12.6/mo
- A self-hosting fallback for the model: GLM-5.3 weights are published on Hugging Face (zai-org/GLM-5.3) under a licence named GLM-5.3, so read its terms before commercial use
- Remote task steering from WeChat, Feishu, or Telegram — keep long-running work moving from any device
Not good for
- Western enterprise teams with geopolitical trust concerns — Z.ai is Beijing-based and subject to Chinese law for cloud API usage
- Security-sensitive codebases — credential handling for remote development over SSH and messaging-platform-triggered tasks needs further review
- Production-critical workflows — Linux support is in beta, and ZCode was absent from Gartner's 2026 Magic Quadrant (not among 12 vendors evaluated)
Our experience
Pricing
Freemium- 2,000 credits per 5 hours, 10,000 per week
- Rolling access to latest flagship models and features
- 20+ coding tools with deep ZCode integration
- 6x Lite usage (12,000 credits per 5 hours, 60,000 per week)
- Day-to-day development on mid-sized repos
- Priority access to latest models
- Curated MCP tools
- Faster generation speeds
- 14x Lite usage (28,000 credits per 5 hours, 140,000 per week)
- Advanced users working on mid-to-large repos
- First access to latest flagship models and features
- Dedicated resources during peak times
Verdict history
Sources
Verification log
- Pricing· No changes
Automated agent
- Pricing· Needs attention
Automated agent
- Profile· Updated
Automated agent
Summary and strengths named GLM-5.2 and MIT-licensed weights; docs list GLM-5.3 and GLM-5.3-Flash and the Hugging Face card names the licence GLM-5.3. Summary, strengths and audiences corrected. Tier credit figures (2,000/12,000/28,000 per 5h) match; Pro and Max list prices and discounted prices not shown on the docs page, so stored values kept.
- Profile· No changes
Automated agent
GLM-5.3, Lite/Pro/Max $12.6/$56/$117.6 discounted, Linux beta all match stored. Stale verdict summary (GLM-5.2) fix already sits on an unpublished draft; not re-drafted. Rating before: conditional, rating after: conditional.
- Profile· No changes
Automated agent
zcode.z.ai/en re-read 2026-09-23: 'Deep GLM-5.3 integration', 'GLM-5.3-Flash built in for multimodal tasks'; Lite $12.60/month, Pro $56/month (6x Lite), Max $117.60/month (14x Lite); macOS, Windows and Linux builds; app version 3.14.3. Matches stored tagline, description and tiers. No licence, MIT or self-hosting statement on the page. Standing self-contradiction unchanged: verdict_summary still says GLM-5.2 and MIT self-hosting; verdict_history is empty, so payload E (draft zcode-glm-5-3-glm-coding-plan-list-prices-rise-max-usage-cut) has still not published. Not duplicated with verdict_set. Linux 'beta' label not visible this read; single read, not changed. No update call, data_hash unchanged.
- Profile· No changes
Automated agent
zcode.z.ai/en re-read 2026-09-18: GLM-5.3 integration with GLM-5.3-Flash built in; Lite $12.60 (list $18), Pro $56 (list $80, 6x Lite), Max $117.60 (list $168, 14x Lite); macOS, Windows, Linux beta. Tagline, description and pricing_tiers all match. No license, MIT or self-hosting statement on the page, and no disclosure about workspace or git-history upload. Known self-contradiction left in place: verdict_summary still says GLM-5.2 and 'MIT-licensed self-hosting'; strengths/audiences also cite MIT-licensed GLM-5.2 weights. Summary fix is staged as payload E on draft zcode-glm-5-3-glm-coding-plan-list-prices-rise-max-usage-cut (verdict_history empty, so not yet published); no duplicate verdict_set. The MIT self-hosting clause in payload E is unsupported by this page. No update call, data_hash unchanged.
- Profile· Updated
Automated agent
Model: stored GLM-5.2 everywhere (tagline, description, strengths, audience, verdict_summary); live page says 'Deep GLM-5.3 integration', 'Optimized for GLM-5.3 across reasoning, coding, and multi-agent collaboration', 'GLM-5.3-Flash built in for multimodal tasks' (read twice, second verbatim). Pricing: live shows list prices struck through at a uniform 30% off: Lite '$12.6/ month$18', Pro '$56/ month$80', Max '$117.6/ month$168'. Stored Pro $72 and Max $160 no longer match the list price (Lite $18 matches; docs.z.ai also says 'Starting at just 18 USD per month'). Usage: stored Pro '5x Lite', Max '20x Lite'; live Pro '6x Lite usage', Max '14x Lite usage', corroborated by docs.z.ai credit table (Lite 2,000/10,000, Pro 12,000/60,000, Max 28,000/140,000 per 5h/week). Description also carried promo prices $16.20/$64.80/$144 that contradicted the entity's own tiers; removed. Platforms added: macOS, Windows, Linux beta. z.ai/subscribe renders client-side (nav only), so pricing_url unreadable; zcode.z.ai used. verdict_summary still says GLM-5.2 and '50-80%' undercut: NOT set here, staged as entity_verdict on draft zcode-glm-5-3-glm-coding-plan-list-prices-rise-max-usage-cut. Comparison titles claude-code-vs-zcode, cursor-vs-zcode, openai-codex-vs-zcode still say GLM-5.2. entity_id from backfill rows.
- Pricing· Updated
Automated agent
Post-promo pricing: Lite $16.20→$18/mo, Pro $64.80→$72/mo, Max $144→$160/mo (~11% increase). 1.5x quota bonus ended Jul 31.
- Pricing· No changes
Automated agent
- Pricing· No changes
Automated agent
- Pricing· No changes
Imported at launch
- Profile· No changes
Imported at launch
Want this running in your business?
We tested this tool and we build with tools like it every day. Tell us the workflow and we will set it up, integrate it and hand it over working.
The conversation runs on Gnosari, one of the tools in this directory. A real conversation, not a sales script.