Hy-MT2 (Hunyuan-MT2)#
Hy-MT2-30B-A3B is Tencent’s translation Mixture-of-Experts language model with 30B total parameters and 3B activated per token. It features 48 transformer layers (layer 0 dense, layers 1–47 MoE), 128 routed experts plus 1 shared expert with top-8 sigmoid routing, Grouped Query Attention (32 Q / 4 KV heads), per-head QK RMSNorm, RoPE, and an in-forward fp32 upcast on the language-model head (enable_lm_head_fp32). It supports a 256K context window.
Task |
Text Generation (MoE, translation) |
Architecture |
|
Parameters |
30B total / 3B activated |
HF Org |
Available Models#
Hy-MT2-30B-A3B: 30B total, top-8 routed experts (out of 128) activated per token, plus 1 shared expert
Architectures#
HyMT2ForCausalLM
Example HF Models#
Model |
HF ID |
|---|---|
Hy-MT2-30B-A3B |
Example Recipes#
Recipe |
Description |
|---|---|
SFT — Hy-MT2-30B-A3B with FSDP2 + EP8 + fp32 LM head |
Try with NeMo AutoModel#
1. Install (NeMo AutoModel):
pip install nemo-automodel
2. Clone the repo to get the example recipes:
git clone https://github.com/NVIDIA-NeMo/Automodel.git
cd Automodel
3. Run the recipe from inside the repo:
automodel --nproc-per-node=8 examples/llm_finetune/hy_mt2/hy_mt2_30b_a3b_sft.yaml
Refer to the NeMo AutoModel Installation Guide and LLM Fine-Tuning Guide.
Fine-Tune the Model#
Refer to the LLM Fine-Tuning Guide and the Large MoE Fine-Tuning Guide.