Category

API

App

PPIO
Date
Price

qwen/qwen3-coder-next

The code intelligence model released by the Qwen team significantly enhances coding capabilities.
Model
LLM
Model capability: function_call
Input:
$0.2/1M tokens
Output:
$1.5/1M tokens

deepseek/deepseek-ocr-2

DeepSeek AI’s multimodal document recognition model achieves high-precision understanding of document structure through semantic causal scanning.
Model
LLM
Model capability: image
Input:
$0.031/1M tokens
Output:
$0.031/1M tokens

moonshotai/kimi-k2.5

Kimi is currently the most versatile and intelligent multimodal model, excelling in agent capabilities, code, and visual understanding.
Model
LLM
Model capability: imageModel capability: videoModel capability: function_call
Input:
$0.572/1M tokens
Output:
$3/1M tokens

zai-org/glm-4.7-flash

The efficient version of the GLM-4.7 series language model launched by Zhipu AI balances performance and efficiency.
Model
LLM
Model capability: thinkingModel capability: function_call
Input:
$0.0715/1M tokens
Output:
$0.429/1M tokens

minimax/minimax-m2.1

Significantly enhances multilingual programming, designed for real-world complex tasks.
Model
LLM
Model capability: function_call
Input:
$0.3/1M tokens
Output:
$1.2/1M tokens

zai-org/glm-4.7

Zhipu’s flagship text model, covering complex task solving and efficient creation across multiple scenarios.
Model
LLM
Model capability: function_call
Input:
$0.572/1M tokens
Output:
$2.286/1M tokens

xiaomimimo/mimo-v2-flash

Xiaomi’s open-source general-purpose language model, adept at programming and reasoning.
Model
LLM
Model capability: function_call
Input:
$0.1/1M tokens
Output:
$0.3/1M tokens

baidu/ernie-4.5-vl-28b-a3b

Baidu’s flagship multimodal large model—a high-performance visual-language understanding engine that supports both thinking and non-thinking modes.
Model
LLM
Model capability: imageModel capability: function_call
Input:
$0.143/1M tokens
Output:
$0.572/1M tokens

zai-org/glm-4.6v

Zhipu High-Performance Visual Understanding Model
Model
LLM
Model capability: imageModel capability: videoModel capability: function_call
Input:
$0.143/1M tokensstarting from
Output:
$0.429/1M tokensstarting from

deepseek/deepseek-v3.2

The open-source large model with a multi-expert mixture-of-experts (MoE) architecture launched by DeepSeek.
Model
LLM
Model capability: function_call
Input:
$0.286/1M tokens
Output:
$0.429/1M tokens

kat-coder

High-performance AI coding model from Kuaishou StreamLake
Model
LLM
Model capability: function_call
Input:
$0.3/1M tokens
Output:
$1.2/1M tokens

zai-org/glm-4.6

Zhipu AI’s flagship language model supports complex agent tasks and high-precision code generation.
Model
LLM
Model capability: thinkingModel capability: function_call
Input:
$0.57/1M tokens
Output:
$2.286/1M tokens