SkyAI

Relay

SkyAI provides an OpenAI-compatible API relay with access to multiple AI models and stable performance.

Qwen3.6 Plus GPT-5 Gemma 3 Gemini 2.5 Flash Lite GPT-4.1 Mini MiniMax M2 DeepSeek V3.1 Qwen2.5 Coder Instruct

SkyAI offers 43 LLM API models.

API pricing per token ranges from $0.010 to $8.80/M (input).

Speed benchmark average: 98 tok/s.

SkyAI is an API aggregator, offering models from multiple vendors.

Avg Speed98.21 tok/s

Latency2.94 s

Updated5/20/2026

Created At12/7/2025

Recharge Rate¥1.00 per $1 quota

Features

DrawingTaskData Export

Login Methods

API Endpoints

Endpoint 1
https://ai.071572.xyz
Endpoint 2
https://riverai.071572.xyz
Endpoint 3
https://asia.071572.xyz
Endpoint 4
http://cn2.071572.xyz
Endpoint 5Historical / Unverified
https://api.071572.xyz
Endpoint 6Historical / Unverified

Supported Models

Model	Speed	Latency	Tests
claude-opus-4-8	43.48 tok/s	5.57s	5
gemini-3.5-flash	231.45 tok/s	3.64s	5
gpt-5.5	54.34 tok/s	2.74s	5
qwen3.6-plus	51.71 tok/s	16.15s	5
gpt-5	168.79 tok/s	1.89s	5
google/gemma-3-27b	46.69 tok/s	2.55s	160
gemini-2.5-flash-lite	276.49 tok/s	1.32s	60
gpt-4.1-mini	99.21 tok/s	2.73s	150
minimax/minimax-m2	63.58 tok/s	12.70s	20
deepseek-ai/DeepSeek-V3.1	67.78 tok/s	3.02s	31
Qwen/Qwen2.5-Coder-7B-Instruct	62.24 tok/s	0.50s	35

Available Models

32 models available (last updated: 5/30/2026)

BAAI/bge-m3siliconflowDeepSeek-R1customDeepSeek-V3.2customDeepSeek-V4-FlashcustomQwen3.5-35B-A3BcustomQwen3.5-397B-A17Bcustombge-m3customclaude-haiku-4-5-20251001vertex-aiclaude-opus-4-6vertex-aiclaude-opus-4-7vertex-aiclaude-opus-4-8customcodex-gpt-image-2customdeepseek-ai/DeepSeek-OCRcustomdeepseek-ai/DeepSeek-R1submodeldeepseek-ai/DeepSeek-V3customgemini-3.1-pro-previewvertex-aigemini-3.5-flashcustomgpt-5codexgpt-5-3customgpt-5-3-minicustom

Showing 20 of 32 models.