AI Proxy Providers · Cheap OpenAI / Claude API

Find AI Proxy Providers for Cheap OpenAI and Claude APIs

Compare AI proxy providers offering OpenAI API proxy, Claude API proxy, and OpenAI-compatible services by pricing rules, supported models, payment methods, and selection guidance.

Quick filters

By model

Site list(32)

kukuai logo

kukuai

Public model pricing is shown per million tokens. Examples include gpt-5.4-mini at ¥0.1950 input / ¥0.8325 output, gpt-5.3-codex at ¥0.3300 input / ¥3.3000 output, claude-sonnet-4-6 at ¥0.8580 input / ¥3.9600 output, and gemini-3.1-pro at ¥1.5400 input / ¥9.2400 output. The only currently confirmed site-credit plans are the ¥6.8 trial plan and the ¥66 site all-access plan; the lowest effective rate comes from the small ¥6.8 trial plan and does not represent one flat unit price across all plans.

$0.01 ≈ $1¥6.830
DeepKey logo

DeepKey

Top-ups are settled at ¥1 = $1 credit, with a ¥5 minimum. Model usage is charged by group multipliers or fixed prices. The comparable cost is recorded from the gpt group, mainly Pro accounts, at 0.2x, or about ¥0.20 per $1 list-price usage. Stable groups recorded here include: gpt 0.2x, gpt-enterprise 0.6x, gpt-azure 1.2x, gpt-image 0.1x; cx2cc 0.2x; gemini 0.3x, gemini-t3 1.8x; claude 0.4x, claude-anti 0.3x, claude-copilot 0.5x, claude-free 0.25x, claude-pro 1.2x, claude-ai / claude-api 3x, claude-aws-platform 2.2x, claude-max 1.8x, claude-max-any-client 1.6x, claude-kiro 0.45x, and claude-windsurf 0.35x. Limited-time, special-price, discount, sale, and out-of-stock-prone groups are excluded.

28.93K$0.02 ≈ $1¥515
FoxCode logo

FoxCode

Usage cards do not expire; the purchase page shows 100M credits for ¥35, 500M for ¥135, 1B for ¥265, and 2B for ¥468. Different model/channel conversion ranges are shown: official CC full-power ¥0.98-2.8 = $1, Turbo ¥1.3-3.5 = $1, Super special CC ¥0.492-1.4 = $1, Ultra special CC ¥0.246-0.7 = $1, Codex channel CC ¥0.123-0.35 = $1, AWS special CC ¥0.03-0.087 = $1, Gemini official ¥0.123-0.35 = $1, Codex full-power official ¥0.123-0.35 = $1, and GPT-Image official ¥0.05-0.07 per image.

177.65K$0.03 ≈ $1¥358
XAI XAPI logo

XAI XAPI

The current featured Codex programming plans support Codex CLI/App/API access and Claude API mapping for Claude Code; plans are valid for 31 days. The lightweight trial is ¥59 for ¥800 card balance, about ¥0.517 per $1; the starter plan is ¥99 for ¥2000 card balance, about ¥0.347 per $1; the ¥198/¥396/¥792 plans provide ¥4000/¥8000/¥16000 card balance, all about ¥0.347 per $1; the enterprise exclusive plan is ¥2000 for ¥100000 card balance, with gpt-image-2, a 20x Pro account, 16 subaccounts, and an enterprise service group, about ¥0.140 per $1.

4.82K$0.05 ≈ $1¥10

Start here

Four things to verify before choosing an AI proxy provider

Start with price, then check pricing rules, model coverage, payment methods, and invoice support. Once the real cost is clear, it is easier to tell which provider is worth it.

Pricing rules

Start with how each provider explains plans, credits, multipliers, and overage pricing. A cheap headline is not enough; record the rules first, then calculate your actual cost.

Model coverage

A proxy may map expensive models to cheaper ones. Verify the model name, protocol compatibility, and response shape before you connect it.

Payment methods

Alipay, WeChat Pay, credit cards, and USDT affect top-up convenience. Teams should also check long-term payment and reimbursement workflows.

Invoice support

For teams, client projects, or company reimbursement, invoice support can matter more than one-off price. The filter only shows providers where invoice support has been observed.

Guides

Claude Code / Codex / Cursor setup guides

Run the smallest working setup before deciding whether to use an AI proxy provider long term.

FAQ

About cheap OpenAI API, Claude API proxies, and proxy services

What is a Claude API proxy?

A Claude API proxy is usually a third-party service that provides Anthropic or Claude-compatible endpoints. Developers use its Base URL and API key to route Claude Code, backend, or automation requests to the target model.

How is an OpenAI API proxy different from the official API?

OpenAI API proxy services usually provide OpenAI-compatible endpoints, separate API keys, prepaid accounts, and custom Base URLs. Differences include pricing rules, available models, stability, payment methods, invoice support, and compatibility.

What should I check before choosing cheap OpenAI API or an AI API proxy?

Prioritize four things: clear pricing rules, support for your payment methods, invoice availability, and coverage for the models and protocols required by Claude, Codex, Cursor, or similar tools.

Is cheap OpenAI API always cheaper than the official API?

Not always. Providers may bill by multipliers, plans, daily credits, balances, or model-specific rules. First confirm the rules, models, and payment terms, then calculate your actual usage cost.

How do I build an AI proxy service?

A typical setup starts with a cloud server, an open-source proxy project, a domain, certificates, proxy configuration, and account authorization. Then users connect with a Base URL and API key.

Read the deployment guide
How does an AI proxy work?

It is a forwarding layer: users send requests to the proxy, and the proxy routes them through its accounts, models, and rules to upstream services. The outside API looks familiar, while forwarding and billing happen in the middle.