Group Detailed Table
API Group Comparison Guide
Group Types and Characteristics
| Group | Type | Rate | Supported Models |
|---|---|---|---|
| default | Hybrid ChatGPT (AZ Channel) + Claude (Reverse Channel) + MJ (Fast Model) + Domestic Models (Doubao + DeepSeek + Qwen) | Official Rate * 1 | OpenAI, Claude, Domestic Models |
| Premium Gemini | Gemini (Google Channel) Models | Official Rate * 1 | Gemini |
| Official Redirect Gemini | Gemini (Google Channel) Models, more expensive (more accounts, higher cost) | Official Rate * 3 | Gemini |
| Pure AZ | Only ChatGPT Models (AZ Channel) + Domestic Models | Official Rate * 1.5 | OpenAI, Claude, Domestic Models |
| Official Redirect | ChatGPT (AZ Channel) + ChatGPT (Official Redirect Channel) + Domestic Models (Use AZ Channel first, fallback to Official Redirect Channel if unavailable) | Official Rate * 3 | OpenAI, Domestic Models |
| Official Redirect OpenAI | ChatGPT Models (Official Redirect Channel) | Official Rate * 6 | OpenAI |
| Premium Official Redirect OpenAI | ChatGPT Models (Official Redirect Channel), more expensive (more accounts) | Rate * 8 | OpenAI |
| Reverse | Supports GPT + Claude + Gemini + Grok | Official Rate * 1.4 | OpenAI, Claude |
| Limited Time Special | Supports Domestic Models (Doubao + DeepSeek + Qwen) | Official Rate * 0.6 | Gemini, Domestic Models |
| Official Redirect Claude 2 | Claude (AWS Official Redirect Channel) | Official Rate * 6 | Claude |
| Official Redirect Claude 3 | Claude (AWS Official Redirect Channel + Anthropic Official Redirect Channel) | Official Rate * 12 | Claude |
| Direct Connect Claude | Claude (Anthropic Official Redirect Channel) | Official Rate * 16 | Claude |
| Claude Code Exclusive | Claude code | Official Rate * 1.5 | Claude Code |
| Claude | Official Rate * 1.4 | Official Rate * 1 | Official Rate * 1.5 |
| Gemini | Official Rate * 1.4 | Official Rate * 1 | Official Rate * 1.5 |
| MJ Painting | None | $0.24 per time | None |
Pros and Cons Comparison
| Features | Reverse | Default (Hybrid) | AZ | Official Redirect | Official Redirect/Premium Official Redirect OpenAI | Premium Gemini | Official Redirect Gemini | Official Redirect Claude | Direct Connect Claude |
|---|---|---|---|---|---|---|---|---|---|
| Pros | Same as ChatGPT official website, smarter, cheaper, but response and API are slightly different, although it is the same model, the official version given might be different | Same as ChatGPT official website, smarter, cheaper | Many channels, sufficient backup, large volume, high concurrency, no moderation, supports FC, TC | High speed, large volume, high concurrency, no moderation, supports FC, TC | Same as ChatGPT official website, high speed, large volume, high concurrency, more accounts, more stable | Same as official website, smarter, cheaper, large volume, stable | Same as official website, smarter, cheaper, large volume, more stable, more accounts | From AWS and Anthropic official, many channels, sufficient backup, large volume, high concurrency | From Anthropic official, more accounts, more stable |
| Cons | Non-official interface | Non-official interface | Microsoft Azure sometimes lags | Affected by official API stability | Affected by official API stability | Affected by official API stability | Higher cost, higher multiplier, affected by official API stability | Affected by official API stability | Affected by official API stability |
| Introduction | All use OpenAI official reverse, strictly forbid frequent health tests like sayl | Reverse, az hybrid, official redirect | Uses enterprise audit-free Azure, supports FC, TC, supports 1000w tpm, uses OpenAI as failure fallback | All use OpenAI official channels, support FC, TC, support 1000w tpm3.5. Uses OpenAI large quota accounts, high concurrency | Uses OpenAI official channels, supports FC, TC, supports 1000w tpm3.5. Uses OpenAI large quota accounts, high concurrency | Google official API, large quota accounts, high concurrency | Google official API, large quota accounts, high concurrency | All use official channels, support function calling, Claude Code, large quota accounts, high concurrency | All use official channels, support function calling, Claude Code, large quota accounts, high concurrency |
Model Support Status
| Model Type | Reverse | Default (Hybrid) | Azure | Official Redirect |
|---|---|---|---|---|
| GPT-4 | Reverse, All supported | Reverse, All supported | Azure, All supported | OpenAI, All supported |
| GPT-3.5 | Reverse, All supported | Reverse, All supported | Azure, All supported | OpenAI, All supported |
| OpenAI Other Basic Models | All supported | All supported | All supported | All supported |
| Midjourney | Not supported | All supported | Not supported | Not supported |
| Domestic Models | Not supported | All supported | Partially supported | Not supported |
| Claude | Reverse, All supported | All supported | All supported | Reverse, All supported |
Notes
You can add tokens on the Token page and add corresponding groups to the token, so as to only call models under that group.
Modified at 2025-12-16 11:52:36