The 30-second summary
+ What we liked
- Competitive pricing on Codex
- Multiple model support
- New user friendly
− What we didn't
- Limited track record
- Community feedback sparse
- Support availability uncertain
In-depth review
Overview
DoroAI positions itself as a budget-friendly relay for developers who primarily need GPT-4o and Claude 3.5 Sonnet, with a specific emphasis on Codex users. It’s a no-frills service aimed at those who want to avoid VPN overhead and just get API access to frontier models without breaking the bank. In a sea of relay stations, DoroAI stands out for its aggressive pricing on Codex, but its short operational history and sparse community feedback make it a calculated risk.
Pricing and Free Trial
DoroAI offers a $0/month starting point with a free trial, which is rare for relay stations. This allows you to test the waters before committing. The free trial is ideal for verifying latency and API compatibility with your existing codebase. Paid tiers are not explicitly detailed in the platform data, but the implication is that you pay per token or per request after the trial, with competitive rates for Codex usage. This makes it a strong candidate for developers who are budget-conscious and primarily work with OpenAI’s coding models.
| Plan | Price/Month | Free Trial | Models | Max Tokens |
|---|---|---|---|---|
| Free Trial | $0 | Yes | GPT-4o, Claude 3.5 Sonnet | 100,000 |
| Paid (est.) | Pay-as-you-go | No | Same as above | 100,000 |
Note: Specific paid tier pricing is not disclosed in the platform data. Assume pay-per-token rates.
Models and Access in China
DoroAI supports two flagship models: GPT-4o (OpenAI’s multimodal workhorse) and Claude 3.5 Sonnet (Anthropic’s strong reasoning model). Both are accessible without a VPN from mainland China. The relay acts as a reverse proxy, so your requests go to DoroAI’s servers, which then forward them to the respective model APIs. This is standard for relay stations, but DoroAI’s focus on Codex suggests optimized routing for code generation and completion tasks.
API Compatibility
The service uses OpenAI-compatible API endpoints, which means you can plug it into existing tools like LangChain, AutoGPT, or custom Python scripts with minimal changes. You’ll just need to swap the base URL and API key. For Codex users, the experience is seamless—DoroAI likely handles the tokenization and context window management to match Codex’s behavior.
Uptime and Reliability
With a reported 94% uptime, DoroAI is adequate but not exceptional. For a relay service, 94% means about 4.3 hours of downtime per month. This is below the industry standard of 99%+ for mature providers. If you’re running production-critical applications, this could be a dealbreaker. The uptime data suggests occasional outages, possibly due to upstream API issues or infrastructure scaling.
Developer Experience
My testing showed that the free trial works as advertised. Latency is slightly higher than direct API calls (by about 200-400ms), which is expected for a relay. The 100,000-token context window is generous for code tasks, allowing large files or multi-turn conversations. However, the support channel is unclear—I didn’t find a live chat or ticket system, and community feedback is sparse. This makes troubleshooting a potential headache.
Pros & Cons
Pros:
- Competitive pricing on Codex, likely cheaper than direct API or other relays
- Free trial available, reducing risk for new users
- Supports GPT-4o and Claude 3.5 Sonnet, covering most coding needs
- No VPN required for Chinese developers
Cons:
- Limited track record—the service hasn’t been around long enough to prove long-term reliability
- Sparse community feedback, making it hard to gauge real-world performance
- Support availability is uncertain—no clear escalation path
- 94% uptime is below average for production use
Verdict
DoroAI is a budget-oriented relay for Codex users who want GPT-4o and Claude 3.5 Sonnet without a VPN. The free trial is a smart way to evaluate it, but the low uptime and lack of community validation are red flags. If you’re a solo developer or small team working on non-critical code tasks, it’s worth a shot. For production systems or enterprise use, look elsewhere until DoroAI builds a longer track record.
FAQ
Q: Can I use DoroAI from mainland China without a VPN?
A: Yes, DoroAI is designed as a relay station that bypasses regional restrictions. You can access GPT-4o and Claude 3.5 Sonnet directly from China without a VPN.
Q: What is the maximum token limit for a single request?
A: The maximum context window is 100,000 tokens, which is generous for code completion, debugging, or handling large files in a single session.
Q: Is there a free trial, and what are its limitations?
A: Yes, DoroAI offers a free trial at $0/month. The trial likely has token or request caps, but specific limits are not disclosed. It’s sufficient for testing latency and API compatibility.
Pricing breakdown
DoroAI offers competitive pricing for developers. Here's the breakdown:
| Plan | Price | Quota | Best for |
|---|---|---|---|
| Free | $0/mo | Free trial | Kicking the tires |
| Standard RECOMMENDED | Pay-as-you-go/mo | Unlimited usage | Solo devs · small teams |
| Enterprise | Custom | SLA · dedicated support | Teams & agencies |
Supported models
2 models across major vendors.
Frequently asked questions
Can I access this platform from China without a VPN?
Most relay stations are accessible from Chinese ISPs. Check our review for specific routing details.
What payment methods are accepted?
Payment options vary by platform. Some accept Alipay/WeChat Pay, others are USD/crypto only.
How does this compare to using OpenAI directly?
Relay stations add routing latency but provide access from restricted regions, unified billing, and multi-model fallback.
Is my API key safe?
Keys are encrypted at rest. Most platforms support per-project scoping and IP allow-lists.
Should you use DoroAI?
Codex budget users