Hunyuan A13B
Active Parameters
80B
Context Length
256K
Modality
Text
Architecture
Mixture of Experts (MoE)
License
Apache 2.0
Release Date
25 Jun 2025
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
System Requirements
VRAM requirements for different quantization methods and context sizes
1,024 tokens
Consumer
9x RTX 4090
24GB VRAM
Datacenter
3x NVIDIA A100
80GB VRAM
Apple Silicon
2x Apple M3 Max
128GB VRAM
256,000 tokens
Consumer
10x RTX 4090
24GB VRAM
Datacenter
3x NVIDIA A100
80GB VRAM
Apple Silicon
2x Apple M3 Max
128GB VRAM
Architecture Diagram
Evaluation Benchmarks
No evaluation benchmarks for Hunyuan A13B available.
Rankings
Overall Rank
-
Coding Rank
-
About Hunyuan A13B
Hunyuan A13B is an open-source 80B Mixture-of-Experts model by Tencent activating 13B parameters for efficient STEM reasoning and agentic tasks. It features dual fast/slow thinking modes and supports quantization down to INT4 across a 256K context.
Technical Specifications
Attention
Attention Structure
Multi-Head Attention
Attention Heads
32
Key-Value Heads
8
Attention Head Dimension
128
Position Embedding
Absolute Position Embedding
RoPE Theta
10,000
Sliding Window Attention
No
Sliding Window Size
-
Sliding Window Ratio
-
Linear Attention
-
Linear Attention Ratio
-
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Auxiliary Parameters
-
Hidden Dimension Size
4,096
Number of Layers
32
FFN Intermediate Size (Dense)
3,072
Multi-Token Prediction Heads
-
Tokenizer
Vocabulary Size
128,167
Mixture of Experts
Total Expert Parameters
13.0B
Number of Experts
65
Active Experts
8
Shared Experts
1
FFN Intermediate Size (per Expert)
3,072
Dense Layers Before MoE
-
About Hunyuan
Tencent Hunyuan large language models with various capabilities.