# Install vLLM editable with our changes
VLLM_USE_PRECOMPILED=1 uv pip install --python .venv/bin/python -e . --torch-backend=auto

# Download model (gated; license must be accepted on huggingface.co)
HF_TOKEN=<token> HF_HUB_ENABLE_HF_TRANSFER=1 \
  .venv/bin/hf download google/translategemma-12b-it

# Serve
VLLM_USE_FLASHINFER_SAMPLER=0 HF_TOKEN=<token> \
  .venv/bin/vllm serve google/translategemma-12b-it \
    --max-model-len 8192 \
    --port 8000 \
    --chat-template-content-format openai

# Translation request
curl -sS http://127.0.0.1:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/translategemma-12b-it",
    "messages": [{
      "role": "user",
      "content": [{
        "type": "text",
        "text": "The quick brown fox jumps over the lazy dog.",
        "source_lang_code": "en",
        "target_lang_code": "es"
      }]
    }],
    "max_tokens": 128,
    "temperature": 0.0
  }'

.venv/bin/python -m pytest \
  tests/entrypoints/test_chat_utils.py::test_extra_fields_preserved_in_text_parts \
  tests/entrypoints/test_chat_utils.py::test_extra_fields_preserved_in_image_parts \
  tests/test_config.py::test_nested_rope_parameters \
  -v

============================= test session starts ==============================
platform linux -- Python 3.12.3, pytest-9.0.3, pluggy-1.6.0 -- /root/vllm/.venv/bin/python
cachedir: .pytest_cache
rootdir: /root/vllm
configfile: pyproject.toml
plugins: anyio-4.13.0, asyncio-1.3.0
asyncio: mode=Mode.STRICT, debug=False, asyncio_default_fixture_loop_scope=None, asyncio_default_test_loop_scope=function
collecting ... collected 3 items

tests/entrypoints/test_chat_utils.py::test_extra_fields_preserved_in_text_parts PASSED [ 33%]
tests/entrypoints/test_chat_utils.py::test_extra_fields_preserved_in_image_parts PASSED [ 66%]
tests/test_config.py::test_nested_rope_parameters PASSED                 [100%]

=============================== warnings summary ===============================
<frozen importlib._bootstrap>:488
  <frozen importlib._bootstrap>:488: DeprecationWarning: builtin type SwigPyPacked has no __module__ attribute

<frozen importlib._bootstrap>:488
  <frozen importlib._bootstrap>:488: DeprecationWarning: builtin type SwigPyObject has no __module__ attribute

.venv/lib/python3.12/site-packages/torch/jit/_script.py:365: 14 warnings
  /root/vllm/.venv/lib/python3.12/site-packages/torch/jit/_script.py:365: DeprecationWarning: `torch.jit.script_method` is deprecated. Please switch to `torch.compile` or `torch.export`.
    warnings.warn(

<frozen importlib._bootstrap_external>:1297
  <frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.

<frozen importlib._bootstrap_external>:1297
  <frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.

-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
======================= 3 passed, 18 warnings in 14.92s ========================

{
  "id": "chatcmpl-b7c34f4f69aa6028",
  "object": "chat.completion",
  "created": 1777861533,
  "model": "google/translategemma-12b-it",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "El rápido zorro marrón salta sobre el perro perezoso.\n",
        "refusal": null,
        "annotations": null,
        "audio": null,
        "function_call": null,
        "tool_calls": [],
        "reasoning": null
      },
      "logprobs": null,
      "finish_reason": "stop",
      "stop_reason": 106,
      "token_ids": null
    }
  ],
  "service_tier": null,
  "system_fingerprint": "vllm-0.1.dev16282+gbade09474-77ae533f",
  "usage": {
    "prompt_tokens": 85,
    "total_tokens": 101,
    "completion_tokens": 16,
    "prompt_tokens_details": null
  },
  "prompt_logprobs": null,
  "prompt_token_ids": null,
  "kv_transfer_params": null
}

=== fr -> en ===
input:  Bonjour, comment allez-vous aujourdhui?
output: Hello, how are you today?

=== en -> ja ===
input:  Where is the train station?
output: 駅はどこですか？

=== cs -> de ===
input:  V nejhorším případě i k prasknutí čočky.
output: Im schlimmsten Fall kann es sogar zum Riss der Linse kommen.