The 30-second summary
+ What we liked
- Dedicated status monitoring page
- Active QQ community
- Available rate acceptable on most channels
- Reverse-proxy Opus 4.6 at ¥2(进)10(出)
− What we didn't
- Slightly above average pricing
- Some channels may have intermittent issues
- Support through QQ group only
In-depth review
IKunCode charges ¥2 per million tokens input and ¥10 per million tokens output for the reverse-proxy Opus 4.6 channel, but for the listed models (GPT-4o, Claude 3.5 Sonnet), you’ll need to check their QQ group for the latest per-model rate card — the platform doesn’t publish public per-token pricing on its status page.
Pricing Breakdown
IKunCode doesn’t offer a free trial tier. You pay upfront via 支付宝 or 微信支付 with no minimum recharge specified. Here’s what we know:
| Model | Input (per million tokens) | Output (per million tokens) | Notes |
|---|---|---|---|
| GPT-4o | Not specified | Not specified | Ask in QQ group |
| Claude 3.5 Sonnet | Not specified | Not specified | Ask in QQ group |
| Opus 4.6 (reverse-proxy) | ¥2 | ¥10 | Separate channel |
The Opus 4.6 pricing is the only concrete number. At ¥10 output, that’s roughly $1.38 USD — cheaper than Anthropic’s direct API rate of $15/million output tokens for Opus. But IKunCode’s GPT-4o and Sonnet pricing is opaque. You’ll need to join their QQ community to get per-model rates.
Cost Comparison vs Direct API
Direct API access from OpenAI (GPT-4o) costs $2.50/input and $10/output per million tokens. Claude 3.5 Sonnet from Anthropic runs $3/input and $15/output. If IKunCode’s markup is similar to the Opus 4.6 channel (roughly 9x cheaper on output), you’re looking at significant savings — but only if you’re using the reverse-proxy Opus 4.6. For GPT-4o and Sonnet, without published rates, you can’t do a direct comparison.
Hidden Fees & Batch Discounts
No batch discounts are advertised. The only “hidden” cost is the lack of transparency: you pay what the QQ group admin tells you. No refund policy is specified, so assume all purchases are final. The 94% uptime means you’ll likely hit 1-2 hours of downtime per day on average — not great for production workloads.
Pros & Cons
Pros
- Dedicated status monitoring page — you can see real-time channel health
- Active QQ community for support and rate updates
- Reverse-proxy Opus 4.6 at ¥2/¥10 is genuinely cheap
- 支付宝 and 微信支付 accepted
Cons
- No public per-token pricing for GPT-4o or Claude 3.5 Sonnet
- 94% uptime is below average for paid relay services
- Support is QQ-only — no ticket system or email
- Some channels have intermittent issues per the platform data
Verdict
IKunCode is a niche pick. If you specifically want cheap Opus 4.6 access and don’t mind joining a QQ group to get rates, it works. But for GPT-4o and Sonnet users, the lack of transparent pricing and below-average uptime (94%) makes it hard to recommend over competitors that publish clear per-million-token rates. The active status page is a nice touch, but it doesn’t compensate for the opacity. Skip this unless you’re already in the QQ community and know exactly what you’re paying.
FAQ
Q: Does IKunCode have a free trial?
A: No. The platform data shows free_trial: False, and no free credits are offered. You must recharge via 支付宝 or 微信支付 before use.
Q: How do I get the actual per-token prices for GPT-4o and Claude 3.5 Sonnet? A: Join their QQ community (link on their status page) and ask in the group. The platform does not publish these rates publicly.
Q: Is the 94% uptime good enough for production use? A: Probably not. 94% uptime means ~1.4 hours of downtime per day. For mission-critical apps, aim for 99%+ uptime relay services. IKunCode is better suited for testing or personal use.
Q: Can I get a refund if a channel is down? A: No refund policy is specified in the platform data. Assume purchases are non-refundable.
Pricing breakdown
IKunCode offers competitive pricing for developers. Here's the breakdown:
| Plan | Price | Quota | Best for |
|---|---|---|---|
| Free | $0/mo | Limited | 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 IKunCode?
Users who want transparency via status page