Skip to content

Models & Pricing

Overview

The Model List page provides a comprehensive catalog of all AI models available on the Wuliang AI platform. You can browse models, compare capabilities, view pricing, and understand rate limits before making API calls. The page supports filtering and searching to help you quickly find the right model for your use case.

Model list page

Prerequisites

  • A registered Wuliang AI account
  • Logged in to the console

Access the Model List

  1. In the console sidebar, navigate to Development > Model List.

    Navigate to model list

  2. The model list page displays all available AI models as cards.

    Model list page loaded

Browse and Filter Models

Use the search bar at the top of the page to filter models by name or model code.

Keyword search

Capability Type Filter

Use the capability type dropdown to filter models by what they can do:

FilterDescription
Text GenerationChat completions, text generation, and conversation models
Image GenerationModels that create images from text prompts
Web SearchSearch-augmented generation models
EmbeddingText embedding and vector representation models
Text-to-Speech (TTS)Models that synthesize speech from text
Speech-to-Text (STT)Models that transcribe audio to text

Capability type filter

Model Cards

Each model is displayed as a card containing the following information:

FieldDescription
Model NameThe display name of the model
Model CodeThe identifier used in API requests (e.g., deepseek-v4-flash)
CapabilityThe type of capability this model provides
Context LengthMaximum number of tokens the model can process in a single request
Input PriceCost per unit for input tokens
Output PriceCost per unit for output tokens
RPM LimitMaximum requests per minute
TPM LimitMaximum tokens per minute

Model card details

Pricing Information

Text Generation Pricing

Text generation models are billed per token. Pricing is displayed per million tokens:

ModelContextInput (per 1M tokens)Output (per 1M tokens)
deepseek-v4-flash1M¥1.0¥2.0
deepseek-v4-pro1M¥12.0¥24.0
qwen3.7-max1M¥2.0¥6.0
glm-5.7200K¥2.0¥5.0
kimi-k2.6256K¥4.0¥12.0
minimax-m31M¥2.0¥8.0

Image Generation Pricing

Image generation models are billed per image:

ModelPrice (per image)
Kolors¥0.01
gpt-image-2¥0.06
nano-banana-2¥0.12
doubao-seedream-4.0¥0.20

Web Search Pricing

Web search is billed per request:

ServicePrice (per request)
web-search¥0.02

Pricing details on model card

Prices shown on the model list page are always up-to-date. For the complete pricing overview, see Pricing.

Selecting a Model

When choosing a model for your use case, consider the following factors:

  1. Capability - Ensure the model supports the type of task you need (text generation, image generation, etc.).
  2. Context Length - Choose a model with sufficient context window for your input data.
  3. Cost - Balance performance requirements with your budget. Smaller models are often sufficient for simpler tasks.
  4. Rate Limits - Ensure the model's RPM and TPM limits meet your throughput requirements.

Model comparison

Notes

  • Model availability and pricing may change. Always refer to the console for the latest information.
  • Some models may be temporarily offline for maintenance. Offline models are indicated with a badge on the model card.
  • Context length represents the maximum combined input and output token count.
  • Rate limits shown are platform defaults. Per-key limits can be configured when creating or editing an API Key.