# Inference config for Qwen2 VL 2B Instruct.
#
# Usage:
#   oumi infer -i -c configs/recipes/vision/qwen2_vl_2b/inference/infer.yaml \
#     --image "tests/testdata/images/the_great_wave_off_kanagawa.jpg"
#
# See Also:
#   - Documentation: https://oumi.ai/docs/en/latest/user_guides/infer/infer.html
#   - Config class: oumi.core.configs.InferenceConfig
#   - Config source: https://github.com/oumi-ai/oumi/blob/main/src/oumi/core/configs/inference_config.py
#   - Other inference configs: configs/**/inference/

# Note, if you are running on a multi-GPU machine, constraint the execution
# to a single GPU, e.g., `export CUDA_VISIBLE_DEVICES=0`

model:
  model_name: "Qwen/Qwen2-VL-2B-Instruct"
  torch_dtype_str: "bfloat16"
  model_max_length: 4096
  chat_template: "qwen2-vl-instruct"
  trust_remote_code: True

generation:
  max_new_tokens: 64
  batch_size: 1

engine: NATIVE