You can deploy this model and start earning money today!

amd/
Llama-3.2-1B-Instruct-FP8-KV

Name: Llama-3.2-1B-Instruct-FP8-KV
Availability: InStock

safetensors

llama

meta-llama/Llama-3.2-1B-Instruct

quantized

llama3.2

fp8

Views

No views yet

Llama-3.2-1B-Instruct-FP8-KV

Introduction
This model was created by applying Quark with calibration samples from Pile dataset.
Quantization Stragegy
- Quantized Layers: All linear layers excluding "lm_head"
- Weight: FP8 symmetric per-tensor
- Activation: FP8 symmetric per-tensor
- KV Cache: FP8 symmetric per-tensor
Quick Start

Download and install Quark
Run the quantization script in the example folder using the following command line:

export MODEL_DIR = [local model checkpoint folder] or meta-llama/Llama-3.2-1B-Instruct 
# single GPU
python3 quantize_quark.py \ 
        --model_dir $MODEL_DIR \
        --output_dir Llama-3.2-1B-Instruct-FP8-KV \                           
        --quant_scheme w_fp8_a_fp8 \
        --kv_cache_dtype fp8 \
        --num_calib_data 128 \
        --model_export quark_safetensors \
        --no_weight_matrix_merge \
        --custom_mode fp8
# If model size is too large for single GPU, please use multi GPU instead.
python3 quantize_quark.py \ 
        --model_dir $MODEL_DIR \
        --output_dir Llama-3.2-1B-Instruct-FP8-KV \                           
        --quant_scheme w_fp8_a_fp8 \
        --kv_cache_dtype fp8 \
        --num_calib_data 128 \
        --model_export quark_safetensors \
        --no_weight_matrix_merge \
        --multi_gpu \
        --custom_mode fp8

Deployment

Quark has its own export format and allows FP8 quantized models to be efficiently deployed using the vLLM backend(vLLM-compatible).

Evaluation

Quark currently uses perplexity(PPL) as the evaluation metric for accuracy loss before and after quantization.The specific PPL algorithm can be referenced in the quantize_quark.py. The quantization evaluation results are conducted in pseudo-quantization mode, which may slightly differ from the actual quantized inference accuracy. These results are provided for reference only.

Evaluation scores

Benchmark	Llama-3.2-1B-Instruct	Llama-3.2-1B-Instruct-FP8-KV(this model)
Perplexity-wikitext2	13.1657	13.4320

amd/Llama-3.2-1B-Instruct-FP8-KV

Llama-3.2-1B-Instruct-FP8-KV

Introduction

Quantization Stragegy

Quick Start

Deployment

Evaluation

Evaluation scores

License

amd/
Llama-3.2-1B-Instruct-FP8-KV