Laozhang API Review: Pricing & Comparison
A well-known, long-established Chinese Claude relay, priced at parity with official rates, with a strong reputation for stability — a solid first choice if you don't want to deal with headaches
Last verified: 2026-08-11 · Visit official site →
Why Laozhang API is many people’s first stop
In Chinese-speaking AI developer communities, if you ask “which domestic Claude relay is good,” Laozhang API (api.laozhang.ai) almost always shows up in the answers. Not because it’s the cheapest, and not because it has the most features, but because it’s pulled off something that sounds simple but isn’t easy at all: operating stably over the long term in an industry where operators vanishing overnight or going down are common occurrences.
This kind of stability matters especially for developers just getting started — switching platforms, re-applying for an API key, testing compatibility — every migration has friction costs. Choosing a reliable starting point is worth more than chasing the absolute lowest price.
The current homepage also leads with an AI image-model online-testing playground (Nano Banana series and GPT Image 2, free/low-price), and the unified API now serves 200+ models covering the full Claude/GPT/Gemini families plus image generation.
What “priced at parity with official rates” actually means
Laozhang API’s core promise is parity pricing with official rates — meaning it bills at Anthropic/OpenAI’s original official pricing, with no markup. Converted at the current exchange rate (roughly ¥7/USD):
| Model | Input price (/M tokens) | Output price (/M tokens) |
|---|---|---|
| Claude Fable 5 | ~¥21 | ~¥105 |
| Claude Opus 4.8 | ~¥35 | ~¥175 |
| Claude Sonnet 4.6 | ~¥21 | ~¥105 |
| Claude Haiku 4.5 | ~¥7 | ~¥35 |
| GPT-5.5 | See official site | See official site |
Note: the prices above are converted from official USD pricing at ¥7/USD — defer to Laozhang API’s official site for actual pricing, since both the exchange rate and official pricing may change over time. New accounts receive a $0.5 test credit for API connectivity verification.
“Priced at parity with official rates” isn’t the same as “the lowest domestic price.” Some relays manage to price below official rates through efficient account-pool management or bulk purchasing — if your monthly call volume is large, the price gap can be significant. See Yunwu’s low-price strategy or 147API’s discounted pricing for reference.
Stability: what is this label actually worth
In the relay market, “stability” is an easy word to say and a hard metric to verify. Laozhang API’s stability credibility mainly comes from two sources:
Long-term community reputation: in domestic user groups for tools like Claude Code and Cursor, negative reviews of Laozhang API are relatively rare, and feedback like “used it for most of a year with no issues” isn’t uncommon.
A sound business model: parity pricing means Laozhang API can’t profit by cutting corners on cost — its revenue mainly comes from economies of scale (fees/service charges accumulated across a large user base). This model provides some positive discipline for long-term operation.
Use cases
The user profile Laozhang API fits best:
- First time using a domestic Claude relay: you don’t want to run into problems, so you pick a community-verified platform to get running first, then optimize later based on your needs
- Individual developers’ daily use: using tools like Claude Code and Cursor every day, needing stable, uninterrupted access, where price isn’t the top priority
- Small teams doing quick validation: during the product-prototyping stage, it’s not worth spending time comparison-shopping and testing — just start with something trustworthy
Scenarios we wouldn’t recommend it for:
- Large-scale calling, with token usage exceeding ¥5,000/month: at this point the price gap becomes significant, and it’s worth specifically comparison-shopping low-price relays
Note: Laozhang API now supports corporate invoicing, service contracts, and monthly settlement (enterprise contact: hi@laozhang.ai) — the earlier “corporate invoicing not confirmed” caveat is outdated, and enterprise users needing compliant procurement can process reimbursements normally.
How to connect
Laozhang API uses an OpenAI-compatible protocol, so the integration is standard:
import anthropic
client = anthropic.Anthropic(
api_key="your Laozhang API key",
base_url="https://api2.laozhang.ai" # API domain migrated to api2.laozhang.ai in 2026-07; api.laozhang.ai still works
)
message = client.messages.create(
model="claude-opus-4-8",
max_tokens=1024,
messages=[{"role": "user", "content": "Hello"}]
)
Claude Code configuration: in the config file under the ~/.claude directory, set ANTHROPIC_API_KEY to your Laozhang API key and ANTHROPIC_BASE_URL to https://api2.laozhang.ai.
Related reviews
- Segmind: Flux/SD image generation from as low as $0.001/image, a go-to for large-batch creative generation
- hvoy.ai: a relay-platform directory and comparison-review hub — not a relay itself, it helps you find the right one
- Sulian AI: low-latency multi-line backup, disaster-recovery architecture, stable production-grade direct connect
- ByteCat: aimed at coding engineers, includes a complete Codex Linux/Windows guide, mainland direct connect, focused on AI coding models
Quick facts
| Pricing model | Pay-as-you-go, priced at parity with official rates (i.e. passed through 1:1 at Anthropic/OpenAI's official pricing, no markup); new accounts get a $0.5 test credit; Claude Opus 4.8 at roughly ¥35/M input, ¥175/M output (converted at the exchange rate) |
|---|---|
| Model coverage | Full Claude/GPT/Gemini coverage, 200+ models via a unified API, plus image models like Nano Banana and GPT Image 2 |
| Latency / SLA | Dedicated mainland direct-connect line, no proxy needed; specific latency isn't published, but community feedback reports good stability with rare large-scale outages |
| Mainland direct connect | Direct connect |
| Best for | Developers / Enterprise |
| Referral program | No public affiliate program found — you can check with the official team. |
Pros
- Established and trustworthy: a long operating history with a solid reputation in the Chinese-speaking Claude user community, reducing the risk of running into problems
- Priced at parity with official rates: billed 1:1 at Anthropic/OpenAI's official pricing with no markup — transparent, trustworthy pricing
- Mainland direct connect: no VPN needed, well suited to domestic teams that need a stable production environment
Cons
- Priced at parity with official rates means no discount: compared to some low-price relays (like Yunwu, 147API), Laozhang API has no price advantage
- No public affiliate program, no referral/rebate mechanism
- The official site's feature set is fairly minimal, with no advanced enterprise-governance features (like usage auditing)
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 →