RinkoAI

RinkoAI provides OpenAI-compatible API endpoints and other model interfaces with a focus on stability and cost-effectiveness.

Categories

Country中国中转站

Qwen3 Coder GPT-OSS Qwen3 Kimi K2 Instruct Gemini 2.0 Flash DeepSeek V3.1 GPT-4.1 Mini Mistral Nemo DeepSeek V3.2

RinkoAI offers 75 LLM API models.

API pricing per token ranges from $0.0050 to $10.00/M (input).

Speed benchmark average: 180 tok/s.

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

Avg Speed179.70 tok/s

Latency1.73 s

Total Tests195

Models14

Updated4/18/2026

Created At8/13/2025

Versionv0.12.14

Online SinceApr 18, 2026, 03:47 AM

Features

DrawingTaskData Export

Website

API Endpoints

Historical / Unverified
https://rinkoai.com
Historical / Unverified
http://rinkoai.com

Login Methods

Passkey

Supported Models

Model	Speed	Latency	Tests
Qwen/Qwen3-Coder-480B	552.48 tok/s	0.89s	5
gpt-oss-120b	335.21 tok/s	1.40s	35
Qwen/Qwen3-235B	234.72 tok/s	1.71s	25
moonshotai/kimi-k2-instruct-0905	217.99 tok/s	0.88s	40
gemini-2.0-flash	113.80 tok/s	6.15s	5
deepseek-ai/DeepSeek-V3.1	102.83 tok/s	1.76s	10
Qwen/Qwen3-30B-A3B	73.72 tok/s	1.37s	10
gpt-4.1-mini	72.66 tok/s	2.60s	10
xai/grok-4-fast-non-reasoning	67.41 tok/s	1.87s	25
claude-4.0-sonnet-search	52.61 tok/s	4.12s	5
MistralAI/Mistral-Nemo	40.81 tok/s	2.15s	10
MistralAI/Mistral-Large	39.11 tok/s	2.12s	5
Qwen/Qwen3-14B	38.30 tok/s	1.19s	5
deepseek-ai/DeepSeek-V3.2-Exp	30.90 tok/s	2.35s	5

Available Models

6 models available (last updated: 4/18/2026)

MiniMaxAI/MiniMax-M2.5customQwen/Qwen3.5-397B-a17Bcustomdeepseek-ai/DeepSeek-V3.1-Terminuscustomdeepseek-ai/DeepSeek-V3.2customgpt-oss-120bcustomzai-org/GLM-5custom

RinkoAI

Features

API Endpoints

Login Methods

Supported Models

Available Models

RinkoAI

Features

API Endpoints

Login Methods

Leaderboard Rankings

Supported Models

Available Models

Leaderboard Rankings