routerhubSI/openrouter-alternative

选用指南

AI 根据 README 整理 · 2026/10/30 02:22:23

适用场景

适合需要单一预付 API 密钥访问多个前沿和开放权重模型，并希望获得明确无审查或 abliterated 模型条目的开发者。

查看 README 依据

<p align="center"><strong>RouterHub is live:</strong> the model catalogue, published prices and API access are at <a href="https://routerhub.si/?utm_source=github&utm_medium=readme&utm_campaign=openrouter-alternative">routerhub.si</a>.</p> --- ## What is this repository? This is a practical **OpenRouter alternative** guide for developers who want: - one prepaid API key across frontier and open-weight models; - explicit **uncensored and abliterated** catalogue entries; - OpenAI- and Anthropic-compatible interfaces where supported; - published model IDs and per-token prices;

> OpenRouter may be the better choice if you need its specific routing features, provider marketplace or integrations. RouterHub is the better fit when you want a simple prepaid key and explicit uncensored/abliterated entries. --- ## Why developers look for an OpenRouter alternative Common reasons: 1. **Predictable prepaid billing.** Top up a balance and pay per token. 2. **Simpler catalogue.** Exact model IDs and published prices in one place.

使用方式

通过 OpenAI 兼容的 Chat Completions API 使用，需要设置 API 密钥和基础 URL，并指定确切的模型 ID。

查看 README 依据

| Qwen 3.8 Max | `qwen3.8-max` | $0.20 | $0.60 | | Qwen 3.8 Max Uncensored | `qwen3.8-max-uncensored` | $0.25 | $0.75 | | GPT-6.1 Sol | `gpt-6.1-sol` | $0.149477 | $0.747385 | | Claude Sonnet 5.5 | `claude-sonnet-5-5` | $0.3587445 | $1.7937225 | Live rates, cached-input prices and the full 37-model catalogue: [routerhub.si/models](https://routerhub.si/models). --- ## Migrating from OpenRouter RouterHub uses OpenAI-compatible Chat Completions for the DeepSeek, GLM, Qwen, Grok and Gemini entries, so migration is usually a base URL, key and model-ID change. ### 1. Create a RouterHub account

Go to [routerhub.si](https://routerhub.si/register?next=onboarding&utm_source=github&utm_medium=readme&utm_campaign=openrouter-alternative) and copy: - **API key** - **Base URL** - the exact **model ID** you want ### 2. Change the client ```python import os from openai import OpenAI client = OpenAI( api_key=os.environ["ROUTERHUB_API_KEY"], base_url=os.environ["ROUTERHUB_BASE_URL"], ) response = client.chat.completions.create( model="MODEL_ID_FROM_DASHBOARD", messages=[ {"role": "user", "content": "Reply with OK."}, ], )

部署要求

需要创建 RouterHub 账户并获取 API 密钥。

查看 README 依据

| Qwen 3.8 Max | `qwen3.8-max` | $0.20 | $0.60 | | Qwen 3.8 Max Uncensored | `qwen3.8-max-uncensored` | $0.25 | $0.75 | | GPT-6.1 Sol | `gpt-6.1-sol` | $0.149477 | $0.747385 | | Claude Sonnet 5.5 | `claude-sonnet-5-5` | $0.3587445 | $1.7937225 | Live rates, cached-input prices and the full 37-model catalogue: [routerhub.si/models](https://routerhub.si/models). --- ## Migrating from OpenRouter RouterHub uses OpenAI-compatible Chat Completions for the DeepSeek, GLM, Qwen, Grok and Gemini entries, so migration is usually a base URL, key and model-ID change. ### 1. Create a RouterHub account

已知限制

OpenRouter 的模型 ID 与 RouterHub 的 ID 不同，不能直接使用。

查看 README 依据

Yes — entries such as `deepseek-v4.1-flash-abliterated` and `glm-5.3-flash-abliterated` are listed in the current catalogue. **How much does it cost?** Published per-token rates are in the [catalogue](https://routerhub.si/models). Pay per token; no subscription. **Can I pay with crypto?** RouterHub accepts card and crypto top-ups. Check the dashboard for current options. **Will my OpenRouter model ID work?** No. Use the exact RouterHub model ID. OpenRouter slugs and RouterHub IDs are different. **Does this repository guarantee ranking or traffic?**