README.md

---
license: mit
---
# Model Overview

- **Model Architecture:** GLM-5
  - **Input:** Text
  - **Output:** Text
- **Supported Hardware Microarchitecture:** AMD MI350/MI355
- **ROCm:** 7.1.0
- **Operating System(s):** Linux
- **Inference Engine:** [vLLM](https://docs.vllm.ai/en/latest/)
- **Model Optimizer:** [AMD-Quark](https://quark.docs.amd.com/latest/index.html) (V0.11.1)
  - **moe**
    - **Weight quantization:** MOE-only, OCP MXFP4, Static
    - **Activation quantization:** MOE-only, OCP MXFP4, Dynamic
- **Calibration Dataset:** [Pile](https://huggingface.co/datasets/mit-han-lab/pile-val-backup)

This model was built with GLM-5 model by applying [AMD-Quark](https://quark.docs.amd.com/latest/index.html) for MXFP4 quantization.

# Model Quantization

The model was quantized from [zai-org/GLM-5](https://huggingface.co/zai-org/GLM-5) using [AMD-Quark](https://quark.docs.amd.com/latest/index.html). The weights and activations are quantized to MXFP4.

**Quantization scripts:**

```python
from quark.torch import LLMTemplate, ModelQuantizer

# --- Register GLM-5 template ---
GLM5_template = LLMTemplate(
    model_type="glm_moe_dsa",
    kv_layers_name=["*kv_a_proj_with_mqa", "*kv_b_proj"],
    q_layer_name="*q_a_proj",
    exclude_layers_name=["lm_head"],
)
LLMTemplate.register_template(GLM5_template)
print(f"[INFO]: Registered template '{GLM5_template.model_type}'")

# --- Configuration ---
model_dir = "zai-org/GLM-5"
output_dir = "amd/GLM-5-MXFP4"
quant_scheme = "mxfp4"
exclude_layers = [
    "*self_attn*",
    "*mlp.gate",
    "*lm_head",
    "*mlp.gate_proj",
    "*mlp.up_proj",
    "*mlp.down_proj",
    "*shared_experts*",
]

# --- Build quant config from template ---
template = LLMTemplate.get("glm_moe_dsa")
quant_config = template.get_config(scheme=quant_scheme, exclude_layers=exclude_layers)

# --- File-to-file quantization (memory-efficient, no full model loading) ---
quantizer = ModelQuantizer(quant_config)
quantizer.direct_quantize_checkpoint(
    pretrained_model_path=model_dir,
    save_path=output_dir,
)

print(f"[INFO]: Quantization complete. Output saved to {output_dir}")

```

# Deployment
### Use with vLLM

This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend.

## Evaluation
The model was evaluated on GSM8K benchmarks. 

### Accuracy

<table>
  <tr>
   <td><strong>Benchmark</strong>
   </td>
   <td><strong>GLM-5 </strong>
   </td>
   <td><strong>GLM-5-MXFP4(this model)</strong>
   </td>
   <td><strong>Recovery</strong>
   </td>
  </tr>
  <tr>
   <td>GSM8K (flexible-extract)
   </td>
   <td>95.45
   </td>
   <td>95.00
   </td>
   <td>99.53%
   </td>
  </tr>
</table>

### Reproduction

The GSM8K results were obtained using the `lm-evaluation-harness` framework, based on the Docker image `rocm/pytorch-private:vllm_glm5_0225`, with vLLM, lm-eval compiled and installed from source inside the image.
The Docker image contains the necessary vLLM code modifications to support this model.

#### Launching server
```
export VLLM_ROCM_USE_AITER=1
export VLLM_ROCM_USE_AITER_FP8BMM=0
export VLLM_ROCM_USE_AITER_FP4BMM=0
vllm serve amd/GLM-5-MXFP4 \
  -tp 8 \
  --block-size 1 \
  --trust-remote-code \
  --max-model-len 4096
```

#### Evaluating model in a new terminal
```
lm_eval \
  --model local-completions \
  --model_args '{"model": "amd/GLM-5-MXFP4", "base_url": "http://localhost:8000/v1/completions", "num_concurrent": 32, "max_retries": 10, "max_gen_toks": 2048, "tokenizer_backend":"None","tokenized_requests":"False" }' \
  --tasks gsm8k \
  --batch_size auto \
  --num_fewshot 5 \
  --trust_remote_code
```

# License
Modifications Copyright(c) 2025 Advanced Micro Devices, Inc. All rights reserved.