mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
250d9c293d
* Update TensorRT-LLM --------- Co-authored-by: Bhuvanesh Sridharan <bhuvan.sridharan@gmail.com> Co-authored-by: Morgan Funtowicz <funtowiczmo@gmail.com> Co-authored-by: Eddie-Wang1120 <wangjinheng1120@163.com> Co-authored-by: meghagarwal <16129366+megha95@users.noreply.github.com>
24 lines
654 B
Python
24 lines
654 B
Python
#!/usr/bin/env python
|
|
import os
|
|
import subprocess
|
|
import sys
|
|
|
|
PROMPT = "Tell a story"
|
|
LLAMA_MODEL_DIR = sys.argv[1]
|
|
EXAMPLES_ROOT = sys.argv[2] if len(sys.argv) > 2 else ""
|
|
LLM_EXAMPLES = os.path.join(EXAMPLES_ROOT, 'llm_examples.py')
|
|
|
|
run_cmd = [
|
|
sys.executable, LLM_EXAMPLES, "--task=run_llm_with_quantization",
|
|
f"--prompt={PROMPT}", f"--hf_model_dir={LLAMA_MODEL_DIR}",
|
|
"--quant_type=int4_awq"
|
|
]
|
|
subprocess.run(run_cmd, check=True)
|
|
|
|
run_cmd = [
|
|
sys.executable, LLM_EXAMPLES, "--task=run_llm_with_quantization",
|
|
f"--prompt={PROMPT}", f"--hf_model_dir={LLAMA_MODEL_DIR}",
|
|
"--quant_type=fp8"
|
|
]
|
|
subprocess.run(run_cmd, check=True)
|