Remove v0 guided decoding

Author: rzabarazesh

.buildkite/test-pipeline.yaml

@@ -128,8 +128,7 @@ steps:
   - tests/entrypoints/offline_mode
   commands:
   - export VLLM_WORKER_MULTIPROC_METHOD=spawn
-  - pytest -v -s entrypoints/llm --ignore=entrypoints/llm/test_lazy_outlines.py --ignore=entrypoints/llm/test_generate.py --ignore=entrypoints/llm/test_generate_multiple_loras.py --ignore=entrypoints/llm/test_guided_generate.py --ignore=entrypoints/llm/test_collective_rpc.py
-  - pytest -v -s entrypoints/llm/test_lazy_outlines.py # it needs a clean process
+  - pytest -v -s entrypoints/llm  --ignore=entrypoints/llm/test_generate.py --ignore=entrypoints/llm/test_generate_multiple_loras.py --ignore=entrypoints/llm/test_guided_generate.py --ignore=entrypoints/llm/test_collective_rpc.py
   - pytest -v -s entrypoints/llm/test_generate.py # it needs a clean process
   - pytest -v -s entrypoints/llm/test_generate_multiple_loras.py # it needs a clean process
   - VLLM_USE_V1=0 pytest -v -s entrypoints/llm/test_guided_generate.py # it needs a clean process

.github/CODEOWNERS

@@ -10,7 +10,6 @@
 /vllm/worker/worker.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
 /vllm/model_executor/layers/sampler.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
 /vllm/model_executor/layers/quantization @mgoin @robertgshaw2-redhat @tlrmchlsmth
-/vllm/model_executor/guided_decoding @mgoin @russellb @aarnphm
 /vllm/multimodal @DarkLight1337 @ywang96
 /vllm/vllm_flash_attn @LucasWilkinson
 /vllm/lora @jeejeelee
@@ -37,7 +36,6 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson
 /tests/entrypoints @DarkLight1337 @robertgshaw2-redhat @simon-mo @aarnphm
 /tests/entrypoints/llm/test_guided_generate.py @mgoin @russellb @aarnphm
 /tests/kernels @tlrmchlsmth @WoosukKwon
-/tests/model_executor/test_guided_processors.py @mgoin @russellb
 /tests/models @DarkLight1337 @ywang96
 /tests/multi_step @alexm-redhat @comaniac
 /tests/multimodal @DarkLight1337 @ywang96

.github/mergify.yml

@@ -149,8 +149,6 @@ pull_request_rules:
       - files=examples/offline_inference/structured_outputs.py
       - files=examples/online_serving/openai_chat_completion_structured_outputs.py
       - files=examples/online_serving/openai_chat_completion_structured_outputs_with_reasoning.py
-      - files~=^vllm/model_executor/guided_decoding/
-      - files=tests/model_executor/test_guided_processors.py
       - files=tests/entrypoints/llm/test_guided_generate.py
       - files~=^tests/v1/structured_output/
       - files=tests/v1/entrypoints/llm/test_guided_generate.py

tests/entrypoints/llm/test_guided_generate.py

@@ -21,7 +21,6 @@
 # which only support regex based constraints in tests.
 GRAMMAR_DECODING_BACKENDS = [
     # (backend, disable_any_whitespace),
-    ("lm-format-enforcer", False),
     ("xgrammar", True),
     ("guidance", True),
 ]

tests/entrypoints/llm/test_lazy_outlines.py

@@ -31,7 +31,7 @@
 # need to change to match the prompt adapter
 PA_NUM_VIRTUAL_TOKENS = 8
 
-GUIDED_DECODING_BACKENDS = ["outlines", "lm-format-enforcer", "xgrammar"]
+GUIDED_DECODING_BACKENDS = ["outlines", "xgrammar"]
 
 
 @pytest.fixture(scope="module")

tests/entrypoints/openai/test_completion.py

@@ -276,7 +276,7 @@ def test_mistral_function_calling(vllm_runner, model: str, dtype: str) -> None:
 
 @pytest.mark.parametrize("model", MODELS)
 @pytest.mark.parametrize("guided_backend",
-                         ["outlines", "lm-format-enforcer", "xgrammar"])
+                         ["outlines", "xgrammar"])
 def test_mistral_guided_decoding(
     monkeypatch: pytest.MonkeyPatch,
     vllm_runner,

tests/model_executor/test_guided_processors.py

@@ -62,7 +62,7 @@ def test_sampling_params_from_request_with_no_guided_decoding_backend(
 
 @pytest.mark.parametrize("request_level_guided_decoding_backend,expected",
                          [("xgrammar", "xgrammar"),
-                          ("lm-format-enforcer", "lm-format-enforcer"),
+                          ("guidance", "guidance"),
                           ("outlines", "outlines")])
 def test_sampling_params_from_request_with_guided_decoding_backend(
         request_level_guided_decoding_backend: str, expected: str,