Repository-wide reasoning
Analyse entire applications with a 1 million token context window on every plan.
Full software lifecycle
Generate code, refactor, debug, document, create tests, review architecture and migrate frameworks.
Ultra-low latency
Instant responses. Dedicated EU infrastructure keeps latency low, so developers stay in flow.
IP protection
Your source code is never used for model training and never leaves the EU - dedicated European infrastructure under your governance.
Developer
- OpenCode
- Claude Code
- Pi
- Other
- Code generation & completion
- Explain, refactor, test
- Multi-file understanding
- Built for developers
- Optimized for speed & quality
- Enterprise grade security
- Best-in-class closed models
- Model choice per request
- Cost & performance optimized
- EU alternative to OpenRouter
- 100% EU Hosting
- GDPR & AI Act Compliant
- Zero Data Retention
- Encrypted in Transit & at Rest
- European Sovereignty
- ChatGPT
- Claude
- Gemini
- Other
All models accessed securely via ZRouter
Read more
Private AI Coding. Built for speed.
ZDev is Zeldoc's private coding AI, running on dedicated European infrastructure. It handles code generation, debugging, refactoring, documentation, tests and architecture guidance - with repository-wide reasoning across up to 1 million tokens of context.
Unlimited usage and fair usage. Max gives unlimited usage under fair use - no per-token billing. Go and Pro include agent credits per month (Pro: 3x Go); when the credits are used up, requests pause until the next month, or continue as passthrough billing if your admin has enabled overage in ZControl. See pricing for the fair usage policy.
ZDev delivers GDPR compliance, AI Act readiness and Zero Data Retention on local/EU models. Note: external frontier models via ZRouter are not covered - data leaves the EU at inference.
Free trial. Organisations can request a 14-day free trial to experience ZDev's speed and context before committing.
Highlights
- Frontier-class coding intelligence
- Multimodal understanding
- Up to 1 million token context
- Ultra-fast, low-latency responses
- Per-minute limit sized for normal coding work - nothing metered per token
- Unlimited usage under fair usage (Max plan)
- 14-day free trial available for organisations
- Dedicated EU infrastructure
Why developers choose ZDev
Key advantages over frontier models
Faster than frontier models
ZDev runs on dedicated European infrastructure optimized for coding workloads, delivering lower latency and faster time-to-first-token than public frontier models. Developers get instant responses without queues or rate limits from shared capacity.
Predictable limits
Each plan has a per-minute request limit sized for normal coding work. Requests above it are answered with a retry signal and go through the next minute - nothing is metered per token, and normal workflows stay well inside it.
Unlimited usage
With the Max plan, developers get unlimited usage within fair use and no per-token billing - focus on building software, not tracking spend. Go and Pro include agent credits at a fixed price per user.
Technology positioning
ZDev vs. frontier models, category by category
| Category | ZDev 1.1 previous model | ZDev 2.0 | Sonnet 4.6 | Sonnet 5.0 | Opus 4.8 | Opus 5.0 | ChatGPT 5.6 Sol | Gemini 3.5 |
|---|---|---|---|---|---|---|---|---|
| Coding Quality | 4.5 | 4.6 | 4.4 | 4.5 | 4.6 | 4.7 | 4.7 | 4.1 |
| Reasoning | 4.2 | 4.6 | 4.1 | 4.1 | 4.6 | 4.7 | 4.7 | 3.8 |
| Speed | 4.8 | 4.9 | 4.0 | 4.0 | 3.8 | 3.8 | 4.1 | 4.2 |
| Multimodal | 4.0 | 4.0 | 4.3 | 4.4 | 4.5 | 4.6 | 4.7 | 4.7 |
| Enterprise Privacy | 5.0 | 5.0 | 2.5 | 2.5 | 2.5 | 2.5 | 2.5 | 2.5 |
| EU Sovereignty | 5.0 | 5.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
| Context | 4.5 | 4.5 | 4.4 | 4.4 | 4.5 | 4.5 | 4.5 | 4.6 |
1-5 scale with one decimal. Higher is better.ZDev 1.1 previous model - runs on dedicated European infrastructureZDev 2.0 - next-generation model with multimodal tuningExternal models graded as used directly (US-hosted)Speed reflects latency from European usersGrades are based on Zeldoc's internal testing and public benchmarks. Coding Quality and Reasoning reflect model capability on software engineering tasks. Speed reflects inference latency experienced by European users - ZDev benefits from dedicated EU infrastructure. Enterprise Privacy reflects data retention policy, training-on-customer-data policy and Cloud Act exposure. EU Sovereignty reflects whether data stays in the EU. Context reflects maximum context window. External frontier models (Claude Sonnet 4.6, Claude Sonnet 5.0, Claude Opus 4.8, Claude Opus 5.0, ChatGPT 5.6 Sol, Gemini 3.5) are US-based; data leaves the EU at inference. ZDev 1.1 and ZDev 2.0 run on dedicated European infrastructure with Zero Data Retention. See Privacy Policy for details.
Fair usage policy
ZDev is designed to give developers unlimited coding power without per-token billing. The full fair usage policy - what counts as fair use, the per-minute limit and what happens beyond included credits - is on the pricing page.