github.com/aws-neuron/neuronx-distributed-inference @v0.7.14366

NeuronX Distributed (NxD) Inference

This package provides a model hub for running inference on Neuronx Distributed (NxD).

Examples

This package includes examples that you can reference when you implement code that uses NxD Inference. * generation_demo.py - A basic generation example for Llama.

Run inference with the inference demo

This package includes an inference demo console script that you can use to run inference. This script includes benchmarking and accuracy checking features that are useful for developers to verify that their models and modules work correctly.

After you install this package, you can run the inference demo with inference-demo. See examples below for how to run the inference demo. You can also run inference_demo --help to view all available arguments.

Example 1: Llama inference with token matching accuracy check

inference_demo \
  --model-type llama \
  --task-type causal-lm \
  run \
    --model-path /home/ubuntu/model_hf/Llama-3.1-8B-Instruct/ \
    --compiled-model-path /home/ubuntu/traced_model/Llama-3.1-8B-Instruct/ \
    --torch-dtype bfloat16 \
    --tp-degree 32 \
    --batch-size 2 \
    --max-context-length 32 \
    --seq-len 64 \
    --on-device-sampling \
    --enable-bucketing \
    --top-k 1 \
    --pad-token-id 2 \
    --prompt "I believe the meaning of life is" \
    --prompt "The color of the sky is" \
    --check-accuracy-mode token-matching \
    --benchmark

Example 2. DBRX inference with logit matching accuracy check

inference_demo \
  --model-type dbrx \
  --task-type causal-lm \
  run \
    --model-path /home/ubuntu/model_hf/dbrx-1layer/ \
    --compiled-model-path /home/ubuntu/traced_model/dbrx-1layer-demo/ \
    --torch-dtype bfloat16 \
    --tp-degree 32 \
    --batch-size 2 \
    --max-context-length 1024 \
    --seq-len 1152 \
    --enable-bucketing \
    --top-k 1 \
    --pad-token-id 0 \
    --prompt "I believe the meaning of life is" \
    --prompt "The color of the sky is" \
    --check-accuracy-mode logit-matching

Example 3. Llama with speculation

inference_demo \
  --model-type llama \
  --task-type causal-lm \
  run \
    --model-path /home/ubuntu/model_hf/open_llama_7b/ \
    --compiled-model-path /home/ubuntu/traced_model/open_llama_7b/ \
    --draft-model-path /home/ubuntu/model_hf/open_llama_3b/ \
    --compiled-draft-model-path /home/ubuntu/traced_model/open_llama_3b/ \
    --torch-dtype bfloat16 \
    --tp-degree 32 \
    --batch-size 1 \
    --max-context-length 32 \
    --seq-len 64 \
    --enable-bucketing \
    --speculation-length 5 \
    --top-k 1 \
    --pad-token-id 2 \
    --prompt "I believe the meaning of life is" \
    --check-accuracy-mode token-matching \
    --benchmark

Example 4. Llama with quantization

inference_demo \
  --model-type llama \
  --task-type causal-lm \
  run \
    --model-path /home/ubuntu/model_hf/Llama-2-7b/ \
    --compiled-model-path /home/ubuntu/traced_model/Llama-2-7b-demo/ \
    --torch-dtype bfloat16 \
    --tp-degree 32 \
    --batch-size 2 \
    --max-context-length 32 \
    --seq-len 64 \
    --on-device-sampling \
    --enable-bucketing \
    --quantized \
    --quantized-checkpoints-path /home/ubuntu/model_hf/Llama-2-7b/model_quant.pt \
    --quantization-type per_channel_symmetric \
    --top-k 1 \
    --pad-token-id 2 \
    --prompt "I believe the meaning of life is" \
    --prompt "The color of the sky is"

Example 5. Llama inference with logit matching accuracy check using custom error tolerances

inference_demo \
  --model-type llama \
  --task-type causal-lm \
  run \
    --model-path /home/ubuntu/model_hf/Llama-2-7b/ \
    --compiled-model-path /home/ubuntu/traced_model/Llama-2-7b-demo/ \
    --torch-dtype bfloat16 \
    --tp-degree 32 \
    --batch-size 2 \
    --max-context-length 32 \
    --seq-len 64 \
    --check-accuracy-mode logit-matching \
    --divergence-difference-tol 0.005 \
    --tol-map "{5: (1e-5, 0.02)}" \
    --enable-bucketing \
    --top-k 1 \
    --pad-token-id 2 \
    --prompt "I believe the meaning of life is" \
    --prompt "The color of the sky is"