The 30-second summary
+ What we liked
- Claude Code focus
- New station with competitive entry pricing
− What we didn't
- Very limited public data and reviews
- Unproven reliability
In-depth review
$0/month is the price tag here — but that’s the free trial, not a sustainable plan.
The Pricing Trap
哈基米 (Hajimi) is a new station targeting developers who want to test the absolute latest models. Their main draw: Claude Code and GPT-5.x access. The free trial sounds good on paper, but here’s the reality check.
The platform doesn’t publish per-token or per-million rates. That’s a red flag for anyone doing cost comparison. You can’t calculate whether it’s cheaper than direct API pricing without actually recharging and testing.
| Pricing Aspect | Details |
|---|---|
| Monthly subscription | $0 (free trial only) |
| Per-token rates | Not disclosed |
| Minimum recharge | Not specified |
| Payment methods | 支付宝, 微信支付 |
| Promo code | None |
Compare this to direct API calls: GPT-4o on OpenAI’s API runs about $2.50 per million input tokens and $10 per million output tokens (2025 pricing). GPT-5.x will likely be more expensive. Without Hajimi’s per-token rates, you’re gambling.
There are no hidden fees disclosed, but there’s also no refund policy. If you recharge ¥100 and the service goes down, you’re stuck.
Pros & Cons
Pros
- Claude Code focus — rare to find a relay station prioritizing this model
- New station means competitive entry pricing to attract users
- Supports local payment (支付宝, 微信支付) — no VPN needed
Cons
- 94% uptime is below average. Most relay stations target 99%+. You’ll hit outages.
- No per-token pricing transparency — can’t compare costs
- Safety rating 3/5 — mediocre. Not the worst, but not trustworthy
- Zero public reviews or community feedback. Unproven reliability.
- No refund policy specified. If they shut down tomorrow, your money’s gone.
Verdict
Skip it for production. The 94% uptime and lack of pricing transparency make this a hard pass for anything serious. If you’re a bleeding-edge tester who absolutely needs GPT-5.x or Claude Code right now and can tolerate downtime, throw a small recharge at it. Otherwise, wait for more data points and uptime guarantees.
The composite score of 56.4/100 tells you everything — this is a high-risk, low-reliability option.
FAQ
Q: Can I use Hajimi without recharging? A: Yes, there’s a free trial at $0/month. But the trial duration and limits aren’t specified. Expect it to be very limited.
Q: How does Hajimi’s pricing compare to OpenAI’s direct API for GPT-5.x? A: Impossible to say — Hajimi doesn’t publish per-token rates. You’d need to recharge and test, which is risky given the no-refund policy.
Q: Is 94% uptime acceptable for a relay station? A: No. Most established stations guarantee 99%+. 94% means ~44 hours of downtime per month. That’s unacceptable for any production workload.
Pricing breakdown
哈基米 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 哈基米?
Bleeding-edge testers only