TokenFly企业方案Enterprise
先测真实负载，
再谈价格和额度Test real load first,
then agree price and capacity
用你的提示词或业务流量做样本测试。价格、并发和产能都来自测试结果，写进合同后在控制台生效。Run a sample test on your own prompts or traffic. Prices, concurrency and capacity come from the results, and take effect in your dashboard once they are in the contract.
{
"provider": "TokenFly",
"api_style": "openai",
"base_url": "https://tokenfly.chat/v1",
"limits": { "requests_per_minute": "…", "max_concurrency": "…" },
"models": [
{ "id": "…", "pricing": { "input_per_million": "…" } }
]
}两种用法Two ways to buy
公开价灵活，随充随用。合同价用承诺量换取，签约后在控制台显示。Public prices are flexible and pay-as-you-go. Contract prices are earned with a commitment and appear in your dashboard after signing.
| 自助使用Self-serve | 企业合同Enterprise contract | |
|---|---|---|
| 价格Price | 公开价Public price | 合同专属价Dedicated contract price |
| 承诺Commitment | 无None | 月度或年度用量Monthly or annual volume |
| 付款Payment | 在线充值，随取随用Online top-up, use any time | 对公付款，按合同结算Bank transfer, settled by contract |
| 额度Capacity | 标准限流Standard limits | 按业务峰值配置Sized to your peak |
| 接入Integration | 复制 Base URL 和 KeyCopy the base URL and key | 一键导入你的平台One-step import into your platform |
| 支持Support | 文档和在线咨询Docs and contact form | 合同约定首次响应时间First-response times in the contract |
把 API 一键导入你的平台Import the API into your platform
企业控制台生成接入包，包含 Base URL、合同内模型、专属价格和限流信息。The enterprise dashboard builds an import package with the base URL, contracted models, your prices and limits.
- 选择 OpenAI 兼容配置、.env 或 LiteLLM 模板Choose OpenAI-compatible JSON, .env or a LiteLLM template
- 创建专用 Key，限定模型、预算和有效期Create a dedicated key scoped by model, budget and expiry
- 下载配置文件，或用短期令牌拉取，Key 不放进 URLDownload the file, or pull it with a short-lived token; keys never go in URLs
- 自动发一次测试请求，显示模型、延迟和结算价格A test request shows the model, latency and billed price
- 模型或合同价变化时，控制台提醒更新配置The dashboard flags when models or prices change
按数据要求选交付形态Pick a form by data requirements
| 形态Form | 交付内容What you get |
|---|---|
| 共享 APIShared API | 共享网关和标准限流Shared gateway, standard limits |
| 专属额度Dedicated quota | 专属并发、模型白名单、合同单价Dedicated concurrency, model allow-list, contract prices |
| 专属算力Dedicated compute | 客户独占的推理集群An inference cluster for you alone |
| 私有化Private | 部署到你的环境，数据不出内网Runs in your environment; data stays on your network |
先用真实业务测试，
再谈签约Test on real work,
then sign
需求沟通Scoping
确认模型、用量、数据要求和上线时间。Agree models, volume, data requirements and timing.
样本测试Sample test
用你的提示词或业务流量做同题对比。Compare on your own prompts or traffic.
评估报告Report
给出质量、延迟、产能和成本结果。Results for quality, latency, capacity and cost.
签约上线Sign and launch
约定价格、额度、支持方式和迁移计划。Agree price, capacity, support and the migration plan.
告诉我们你的业务规模Tell us your scale
带上你的模型和预估用量，我们会先判断是否适合，再安排样本测试。Bring your models and expected volume. We will check the fit first, then set up a sample test.