GLM-4#
GLM-4 is Tsinghua University (THUDM)’s fourth-generation General Language Model, featuring strong multilingual capabilities and tool-use support.
Task |
Text Generation |
Architecture |
|
Parameters |
9B – 32B |
HF Org |
Available Models#
GLM-4-9B-Chat-HF (
GlmForCausalLM): 9BGLM-4-32B-0414 (
Glm4ForCausalLM): 32B
Architectures#
GlmForCausalLM— GLM-4 seriesGlm4ForCausalLM— GLM-4-0414 series
Example HF Models#
Model |
HF ID |
|---|---|
GLM-4-9B-Chat-HF |
|
GLM-4-32B-0414 |
Example Recipes#
Recipe |
Description |
|---|---|
SFT — GLM-4 9B on SQuAD |
|
SFT — GLM-4 9B on HellaSwag with FP8 |
Try with NeMo AutoModel#
1. Install (full instructions):
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/glm/glm_4_9b_chat_hf_squad.yaml
Run with Docker
1. Pull the container and mount a checkpoint directory:
docker run --gpus all -it --rm \
--shm-size=8g \
-v $(pwd)/checkpoints:/opt/Automodel/checkpoints \
nvcr.io/nvidia/nemo-automodel:26.02.00
2. Navigate to the AutoModel directory (where the recipes are):
cd /opt/Automodel
3. Run the recipe:
automodel --nproc-per-node=8 examples/llm_finetune/glm/glm_4_9b_chat_hf_squad.yaml
See the Installation Guide and LLM Fine-Tuning Guide.
Fine-Tuning#
See the LLM Fine-Tuning Guide.