Tencent TokenHub · emploidai Marketplace
emploidai Marketplace
Add-onsAppletsPlugins
Search tools, teams, and capabilitiesPublish
MarketplacePluginsTencent TokenHub
Plugin
Limited listing

Tencent TokenHub

by lws123321 · v0.0.4

Models provided by Tencent TokenHub, including DeepSeek, Hunyuan, GLM, Kimi, MiniMax series.

766 installsUpdated Jun 4, 2026
Publisher information is incomplete

This community listing does not yet include every recommended support, privacy, pricing, and permission disclosure. Review the available package permissions before installing.

Capabilities

Models

Available inside your emploidai workspace after installation.

Data sources

Available inside your emploidai workspace after installation.

Category

model

Version

0.0.4lws123321

Requirements

Maximum memory 256MB

Pricing

Not disclosed by publisher

Security & access

Review before installing

CompatibleRequires emploidai 1.0.0+

Permissions

  • Uses model capability

Dependencies

No additional dependencies

Resources

Privacy policy
emploidai Marketplace

Discover capabilities. Review access. Install inside your workspace.

DocumentationSecuritySupportPrivacyTerms

Tencent TokenHub Dify Plugin

Tencent TokenHub plugin for Dify. It is compatible with the OpenAI API protocol and supports a wide range of LLMs.

Overview

  • Plugin Name: Tencent TokenHub
  • Provider Type: LLM
  • Runtime: Python 3.12
  • Authentication: API Key (custom API Base URL is optional)

Supported Models

Model Namemodel ValueCapabilities
Hy3 previewhy3-previewTool calling, thinking toggle, reasoning effort
HY 2.0 Thinkhunyuan-2.0-thinking-20251109Reasoning model (thinking always on)
HY 2.0 Instructhunyuan-2.0-instruct-20251111Instruction model
Hunyuan-rolehunyuan-role-latestRole-play
DeepSeek-V4-Prodeepseek-v4-proTool calling, vision, thinking toggle, reasoning effort
DeepSeek-V4-Flashdeepseek-v4-flashTool calling, thinking toggle, reasoning effort
DeepSeek-V3.2deepseek-v3.2Tool calling, thinking toggle, reasoning effort
DeepSeek-V3.1deepseek-v3.1-terminusTool calling, thinking toggle
DeepSeek-R1-0528deepseek-r1-0528Reasoning model
DeepSeek-V3-0324deepseek-v3-0324Tool calling
GLM-5.1glm-5.1Tool calling, thinking toggle
GLM-5V-Turboglm-5v-turboVision, tool calling, thinking toggle
GLM-5-Turboglm-5-turboTool calling, thinking toggle
GLM-5glm-5Tool calling, thinking toggle
Kimi-K2.6kimi-k2.6Tool calling, thinking toggle
Kimi-K2.5kimi-k2.5Tool calling, thinking toggle
MiniMax-M2.7minimax-m2.7Tool calling (thinking always on)
MiniMax-M2.5minimax-m2.5Tool calling (thinking always on)

Thinking toggle is forwarded to TokenHub as thinking: {"type": "enabled" | "disabled"}. Reasoning effort is forwarded as reasoning_effort: low | medium | high and only takes effect when thinking is enabled. See the TokenHub thinking docs for details.

Installation & Usage

  1. Create an API Key in the Tencent TokenHub Console.
  2. Install this plugin from the Dify plugin marketplace or via the local plugin installation entry.
  3. Fill in the plugin configuration:
    • API Key (required)
    • API Base URL (optional, default: https://tokenhub.tencentmaas.com/v1)
  4. Select any supported model to start.

Configuration

  • api_key: Tencent TokenHub API Key, used for authentication.
  • api_base: OpenAI-compatible API base URL. Default:
    • https://tokenhub.tencentmaas.com/v1

Troubleshooting

  • 401 Unauthorized: API Key is invalid, expired, or contains extra spaces.
  • 404 / model not found: Make sure the model ID matches the built-in model list.
  • Request timeout: Check network connectivity, or retry with another model.
  • Error when using a custom api_base: Make sure the URL includes /v1 and uses HTTPS.

API Documentation

  • Text Generation

Privacy

This plugin does not persist user prompts, model outputs, or API keys on the plugin side. To complete model inference, request content will be sent to Tencent TokenHub servers for processing. See PRIVACY.md for details.