T
Hunyuan 2.0 Instruct
FrontierTencent•Released on 2025-12-05
Tencent's Hunyuan 2.0 Instruct model is optimized for natural chat, creative writing, and business Q&A scenarios. Built on MoE architecture with 406B total parameters (32B active), it supports 256K context and excels in high-concurrency applications requiring fast responses. Best for instruction following and conversational AI.
83
Overall Score
Core Specs
256K
Context Window
32K
Max Output
ReasoningOpen Sourcetext
Scenario Scores
Pros & Cons
Sentiment0% +0% ·0% −
Pros
- +256K context window
- +Optimized for chat and instruction following
- +Fast response for high concurrency
- +Strong creative writing ability
- +Integrated into Tencent ecosystem
Cons
- −Recent 463% price increase (March 2026)
- −Less capable for complex reasoning vs Think version
- −Limited availability outside China
- −Not open source
Pricing
Input (per 1M tokens)$0.63
Output (per 1M tokens)$1.55
Free trial available
Updated on 2026-03-12
Get Started
1Visit the provider's website
2Create an account
3Start using the model
Benchmarks
mmlu86.2%
humanEval82.5%