Site descriptions, notices, and reviews are shown in the language in which they were submitted.
稳定接入 SOTA 模型。 更优的模型价格。 Modelflare 为 AI 工具、应用和 Agent 提供稳定的模型请求服务，聚合多类前沿 SOTA 模型，并提供清晰的价格、用量与请求状态。
Price listing
Fable 5
5 public records are listed; open the full page to review details as needed.
Price listing
Opus 4.8
5 public records are listed; open the full page to review details as needed.
Price listing
Sonnet 5
5 public records are listed; open the full page to review details as needed.
Price listing
GPT 5.6 Sol
4 public records are listed; open the full page to review details as needed.
Price listing
GPT 5.6 Terra
4 public records are listed; open the full page to review details as needed.
Price listing
GPT 5.5
4 public records are listed; open the full page to review details as needed.
Price listing
GPT Image 2
1 public records are listed; open the full page to review details as needed.
Current public prices are shown by model and channel. Token prices use CNY per 1M tokens unless another billing unit is shown.
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| claude-first-topup | claude-fable-5 | 输入 ¥2.02 / 输出 ¥10.1 / 缓存 ¥0.2019 / 写入 ¥4.04 | 0% | — | — | — |
| claude-tmp-cheap | claude-fable-5 | 输入 ¥4.04 / 输出 ¥20.19 / 缓存 ¥0.4038 / 写入 ¥8.08 | 0% | — | — | — |
| claude-award | claude-fable-5 | 输入 ¥16.82 / 输出 ¥84.13 / 缓存 ¥1.68 / 写入 ¥33.65 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| claude-first-topup | claude-opus-4-8 | 输入 ¥1.01 / 输出 ¥5.05 / 缓存 ¥0.1009 / 写入 ¥2.02 | 0% | — | — | — |
| claude-tmp-cheap | claude-opus-4-8 | 输入 ¥2.02 / 输出 ¥10.1 / 缓存 ¥0.2019 / 写入 ¥4.04 | 0% | — | — | — |
| claude-award | claude-opus-4-8 | 输入 ¥8.41 / 输出 ¥42.06 / 缓存 ¥0.8413 / 写入 ¥16.82 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| claude-first-topup | claude-sonnet-5 | 输入 ¥0.4038 / 输出 ¥2.02 / 缓存 ¥0.0404 / 写入 ¥0.8076 | 0% | — | — | — |
| claude-tmp-cheap | claude-sonnet-5 | 输入 ¥0.8076 / 输出 ¥4.04 / 缓存 ¥0.0808 / 写入 ¥1.62 | -1% | — | — | — |
| claude-award | claude-sonnet-5 | 输入 ¥3.37 / 输出 ¥16.82 / 缓存 ¥0.3365 / 写入 ¥6.73 | -1% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| openai-first-topup | gpt-5.6-sol | 输入 ¥0.5048 / 输出 ¥3.03 / 缓存 ¥0.0505 / 写入 ¥0.6309 | +0.4% | — | — | — |
| openai-award | gpt-5.6-sol | 输入 ¥1.01 / 输出 ¥6.06 / 缓存 ¥0.1009 / 写入 ¥1.26 | +0.4% | — | — | — |
| openai-stable | gpt-5.6-sol | 输入 ¥2.69 / 输出 ¥16.15 / 缓存 ¥0.2692 / 写入 ¥3.37 | +0.4% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| openai-first-topup | gpt-5.6-terra | 输入 ¥0.2019 / 输出 ¥1.21 / 缓存 ¥0.0202 / 写入 ¥0.2524 | -1% | — | — | — |
| openai-award | gpt-5.6-terra | 输入 ¥0.4038 / 输出 ¥2.42 / 缓存 ¥0.0404 / 写入 ¥0.5048 | -20% | — | — | — |
| openai-stable | gpt-5.6-terra | 输入 ¥1.08 / 输出 ¥6.46 / 缓存 ¥0.1077 / 写入 ¥1.35 | +114% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| openai-first-topup | gpt-5.5 | 输入 ¥0.5048 / 输出 ¥3.03 / 缓存 ¥0.0505 / 写入 ¥0 | 0% | — | — | — |
| openai-award | gpt-5.5 | 输入 ¥1.01 / 输出 ¥6.06 / 缓存 ¥0.1009 / 写入 ¥0 | 0% | — | — | — |
| openai-stable | gpt-5.5 | 输入 ¥2.69 / 输出 ¥16.15 / 缓存 ¥0.2692 / 写入 ¥0 | 0% | — | — | — |
| Channel | Provider model | Current price | Price trend | Uptime | Fake % | Latency |
|---|---|---|---|---|---|---|
| Media | gpt-image-2 | 输入 ¥25.37 / 输出 ¥152.23 / 缓存 ¥6.34 / 写入 ¥0 | +567% | — | — | — |
Welcome to Modelflare Modelflare’s first priority is stability. We help more users access global SOTA AI models reliably and at lower cost. Credits are denominated in US dollars: top up US$1 and US$1 is added to your credit balance. New users receive US$0.50 automatically after signup. If you run into request failures, slow responses, unavailable models, top-up issues, or billing questions, please contact us through support tickets, our after-sales QQ group, or Telegram.
🎉 The grok-award group is now available! Demand for Grok has been growing, so we’ve added the new grok-award group. Its multiplier is just 0.03, with pricing as low as 3% of official list prices. The group is now available for selection. Choose grok-award when creating or updating an API Key to get started.
Kimi K3 special-pricing availability update Kimi K3 special pricing is temporarily unavailable because of upstream compute constraints. Availability and stability are not guaranteed and may change without notice. Check the pricing page for the currently available groups before use. Do not use Kimi K3 for production or critical workloads.
ModelFlare API Key Fallback Groups Are Now Available Configure ordered fallback groups under API Keys. Requests try channels in the primary group first, then switch through fallback groups. Billing follows the group that serves the request.
ModelFlare Top-up Special is now available. Any single top-up reaching US$20 unlocks the openai-first-topup group. The current rate is 0.015x. Create or update an API key and select openai-first-topup to use it.
3.5 / 5 from 15 approved reviews
ui挺好看的，感觉不像是国内的站点，速度还可以
你们模型收费美元的，怎么这么贵呢？
在站点领了点鸡蛋，用起来还不错
用了一段时间感觉挺良心的，解决问题回复速度很快，然后交流也很真诚，前段时间反馈缓存问题，然后就上了一个缓存保障
试试少充了一点，用的特惠分组，跑起来整体还可以，比较稳定
还行，最近很多站点都炸了，这里还能用，可以了
0.015分组表现力不输豆包
Uptime is the share of recent probes where the service was available; a higher value generally indicates better stability.
Model consistency uses model pass results to show whether the endpoint behaves like the target model.
Pricing and latency come from public monitoring data and are intended for side-by-side gateway comparison.
Data is provided for reference. Third-party gateways may experience outages or balance loss, so test first and add only a small balance when needed.