GPTAPI.US Review: Pricing & Comparison
Dual US-China regions, PayPal + Alipay dual-currency payment, stable direct connect to GPT/Claude/Gemini
Last verified: 2026-06-29 · Visit official site →
PayPal settlement: a rare option among domestic relays
The vast majority of domestic AI API relays only support RMB settlement via Alipay/WeChat Pay, which is a barrier for overseas users, developers who need USD billing cycles, or those who need USD reimbursement documentation.
GPTAPI.US (www.gptapi.us) is one of the few platforms offering both PayPal (USD) and Alipay (RMB) settlement, serving two types of users:
- Domestic developers: Alipay settlement, mainland direct connect, quick to get started
- Overseas Chinese developers or those with overseas accounts: PayPal settlement, USD billing, convenient for corporate reimbursement
With nodes deployed in both the US and China regions, developers in both locations should, in theory, get relatively low-latency access.
Pricing
Pay-as-you-go, no monthly fee, with dual-currency settlement support:
| Settlement method | Target users | Currency |
|---|---|---|
| Alipay | Domestic developers | RMB |
| PayPal / Credit card | Overseas Chinese developers, teams with overseas accounts | USD |
The primary promoted model is GPT-4o-mini (OpenAI’s lightweight model), which stands out on cost-effectiveness for high-frequency, low-cost tasks like text classification, content summarization, and simple Q&A. Check the official site for live per-model pricing.
Model coverage and integration
Focused on three major overseas vendors:
- OpenAI/GPT: GPT-5.5, GPT-4o, GPT-4o-mini (core promoted model), the o-series
- Anthropic/Claude: Claude Fable 5, Opus 4.8, Sonnet 4.6
- Google/Gemini: Gemini 3.5 Flash/Pro
Doesn’t emphasize domestic open-source models like DeepSeek/Qwen. Sufficient for scenarios that only need core APIs from the three major vendors.
# Connecting to GPTAPI.US
export OPENAI_BASE_URL="https://api.gptapi.us/v1"
export OPENAI_API_KEY="your Key"
# Or for Claude Code
export ANTHROPIC_BASE_URL="https://api.gptapi.us"
export ANTHROPIC_API_KEY="your Key"Mainland direct connect, no proxy needed. Dual US-China regional nodes provide low latency in both locations. Check real-world testing for specific latency figures — no public SLA numbers are available.
Who should choose GPTAPI.US
The most typical use cases:
Mixed US-China teams: teams with both domestic members (Alipay) and overseas members (PayPal) can manage everything through a single platform, with each member choosing their preferred payment method.
Enterprises needing USD reimbursement documentation: some corporate finance departments require USD documentation for reimbursement — PayPal settlement satisfies this requirement, which a pure-RMB platform can’t.
Heavy GPT-4o-mini usage: for high-frequency, low-cost scenarios like content classification, summarization, and simple Q&A, GPT-4o-mini’s per-unit price advantage adds up significantly at scale.
Not a good fit for: purely domestic scenarios chasing rock-bottom prices (mid-range pricing here); needs for open-source models like DeepSeek/Qwen; needs for multimodal (image/video) capability.
If you need additional access to Grok (xAI) models, V-API is one of the few domestic direct-connect relays covering Grok, and it also supports Claude/GPT/Gemini. If you already have a proxy environment and need broader routing across both domestic and global models, FlowBar’s dual-coverage plan includes domestic models like DeepSeek/Qwen and is worth a look. If you want a zero-change integration (just swap the base_url) and prioritize low-latency mainland direct connect, MoleAPI’s under-50ms direct-connect plan has a similar positioning to GPTAPI.US.
Information verified 2026-06-29. Dual-currency payment methods and specific model pricing are subject to www.gptapi.us’s official site.
Related reviews
- IKunCode: coding-focused relay, active QQ community, pay-as-you-go with no plans, GPT-5.5 as low as ¥1/6 (per million tokens)
- Inworld Router: 200+ models at 0% markup, deeply optimized for gaming and entertainment AI use cases, smart routing by request complexity
- Yinhe API: $0.4 signup bonus, dedicated Claude Code optimization, a good pick for lightweight testing
- Segmind: Flux/SD image generation from as low as $0.001/image, a top choice for high-volume creative generation
Quick facts
| Pricing model | Pay-as-you-go, no monthly fee; dual US-China regional service, supports PayPal (USD) and Alipay (RMB) payment |
|---|---|
| Model coverage | GPT/Claude/Gemini mainstream overseas models, with dual US-China regional coverage |
| Latency / SLA | Dual US-China regional nodes, mainland direct connect; check real-world testing for specific latency |
| Mainland direct connect | Direct connect |
| Best for | Developers / Enterprise |
| Referral program | No public affiliate program found — check with the official team to confirm. |
Pros
- Dual US-China regional service, catering to both domestic developers (Alipay/mainland direct connect) and overseas Chinese developers (PayPal/credit card)
- Supports USD settlement via PayPal — a rare international payment option among comparable domestic relays, convenient for developers with overseas accounts
- GPT-4o-mini, a cost-effective model, is the core promoted option, well suited to high-volume, low-cost text processing scenarios
Cons
- Model coverage is relatively concentrated on the three major vendors — GPT/Claude/Gemini — without emphasis on domestic models like DeepSeek
- Mid-range pricing, not suited to extremely price-sensitive use cases
- Limited public data on latency and stability
Compare more AI API relays
See the full comparison board — filter by price tier, model coverage, and mainland direct-connect status.
Back to the comparison board →