output = replicate.run(
    "medicapture/llama2-assistant:0d60e02f269cc8aff0ac612b27a8efd18bbce6356dfe60b825fff014222cab61",
    input={
        "debug": False,
        "top_k": 50,
        "top_p": 0.9,
        "temperature": 0.75,
        "system_prompt": "You are a helpful assistant.",
        "max_new_tokens": 128,
        "min_new_tokens": -1
    }
)

# The medicapture/llama2-assistant model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/medicapture/llama2-assistant/api#output-schema
    print(item, end="")

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "medicapture/llama2-assistant:0d60e02f269cc8aff0ac612b27a8efd18bbce6356dfe60b825fff014222cab61",
    "input": {
      "debug": false,
      "top_k": 50,
      "top_p": 0.9,
      "temperature": 0.75,
      "system_prompt": "You are a helpful assistant.",
      "max_new_tokens": 128,
      "min_new_tokens": -1
    }
  }' \
  https://api.replicate.com/v1/predictions