A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 08:30 UTC
[Doc] Update references to doc files (#18637)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Cyrus Leung committed
May 24, 2025 at 06:49 UTC
7d9216495c389410c2901084336239bc626611d5
13 files changed
+16
-19
.github/mergify.yml
+2
-4
@@ -58,7 +58,7 @@ pull_request_rules:
58
- files~=^benchmarks/structured_schemas/
59
- files=benchmarks/benchmark_serving_structured_output.py
60
- files=benchmarks/run_structured_output_benchmark.sh
61
- - files=docs/source/features/structured_outputs.md
61
+ - files=docs/features/structured_outputs.md
62
- files=examples/offline_inference/structured_outputs.py
63
- files=examples/online_serving/openai_chat_completion_structured_outputs.py
64
- files=examples/online_serving/openai_chat_completion_structured_outputs_with_reasoning.py
@@ -135,9 +135,7 @@ pull_request_rules:
135
- files~=^tests/entrypoints/openai/tool_parsers/
136
- files=tests/entrypoints/openai/test_chat_with_tool_reasoning.py
137
- files~=^vllm/entrypoints/openai/tool_parsers/
138
- - files=docs/source/features/tool_calling.md
139
- - files=docs/source/getting_started/examples/openai_chat_completion_client_with_tools.md
140
- - files=docs/source/getting_started/examples/chat_with_tools.md
138
+ - files=docs/features/tool_calling.md
139
- files~=^examples/tool_chat_*
140
- files=examples/offline_inference/chat_with_tools.py
141
- files=examples/online_serving/openai_chat_completion_client_with_tools_required.py
docker/Dockerfile
+2
-2
@@ -2,8 +2,8 @@
2
# to run the OpenAI compatible server.
3
4
# Please update any changes made here to
5
-# docs/source/contributing/dockerfile/dockerfile.md and
6
-# docs/source/assets/contributing/dockerfile-stages-dependency.png
5
+# docs/contributing/dockerfile/dockerfile.md and
6
+# docs/assets/contributing/dockerfile-stages-dependency.png
7
8
ARG CUDA_VERSION=12.8.1
9
#################### BASE BUILD IMAGE ####################
docs/contributing/overview.md
+2
-3
@@ -130,9 +130,8 @@ The PR needs to meet the following code quality standards:
130
understand the code.
131
- Include sufficient tests to ensure the project stays correct and robust. This
132
includes both unit tests and integration tests.
133
-- Please add documentation to `docs/source/` if the PR modifies the
134
- user-facing behaviors of vLLM. It helps vLLM users understand and utilize the
135
- new features or changes.
133
+- Please add documentation to `docs/` if the PR modifies the user-facing behaviors of vLLM.
134
+ It helps vLLM users understand and utilize the new features or changes.
135
136
### Adding or Changing Kernels
137
docs/source/assets/contributing/dockerfile-stages-dependency.png
Binary files a/docs/source/assets/contributing/dockerfile-stages-dependency.png and /dev/null differ
tools/update-dockerfile-graph.sh
+1
-1
@@ -24,7 +24,7 @@ if printf '%s\n' "${FILES[@]}" | grep -q "^docker/Dockerfile$"; then
24
fi
25
26
# Define the target file path
27
- TARGET_GRAPH_FILE="docs/source/assets/contributing/dockerfile-stages-dependency.png"
27
+ TARGET_GRAPH_FILE="docs/assets/contributing/dockerfile-stages-dependency.png"
28
29
# Ensure target directory exists
30
mkdir -p "$(dirname "$TARGET_GRAPH_FILE")"
vllm/config.py
+2
-2
@@ -988,7 +988,7 @@ class ModelConfig:
988
self.use_async_output_proc = False
989
return
990
991
- # Reminder: Please update docs/source/features/compatibility_matrix.md
991
+ # Reminder: Please update docs/features/compatibility_matrix.md
992
# If the feature combo become valid
993
from vllm.platforms import current_platform
994
if not current_platform.is_async_output_supported(self.enforce_eager):
@@ -1004,7 +1004,7 @@ class ModelConfig:
1004
if self.runner_type == "pooling":
1005
self.use_async_output_proc = False
1006
1007
- # Reminder: Please update docs/source/features/compatibility_matrix.md
1007
+ # Reminder: Please update docs/features/compatibility_matrix.md
1008
# If the feature combo become valid
1009
if speculative_config:
1010
self.use_async_output_proc = False
vllm/engine/arg_utils.py
+1
-1
@@ -1084,7 +1084,7 @@ class EngineArgs:
1084
disable_log_stats=self.disable_log_stats,
1085
)
1086
1087
- # Reminder: Please update docs/source/features/compatibility_matrix.md
1087
+ # Reminder: Please update docs/features/compatibility_matrix.md
1088
# If the feature combo become valid
1089
if self.num_scheduler_steps > 1:
1090
if speculative_config is not None:
vllm/engine/output_processor/multi_step.py
+1
-1
@@ -67,7 +67,7 @@ class MultiStepOutputProcessor(SequenceGroupOutputProcessor):
67
@staticmethod
68
@functools.lru_cache
69
def _log_prompt_logprob_unsupported_warning_once():
70
- # Reminder: Please update docs/source/features/compatibility_matrix.md
70
+ # Reminder: Please update docs/features/compatibility_matrix.md
71
# If the feature combo become valid
72
logger.warning(
73
"Prompt logprob is not supported by multi step workers. "
vllm/platforms/cpu.py
+1
-1
@@ -75,7 +75,7 @@ class CpuPlatform(Platform):
75
import vllm.envs as envs
76
from vllm.utils import GiB_bytes
77
model_config = vllm_config.model_config
78
- # Reminder: Please update docs/source/features/compatibility_matrix.md
78
+ # Reminder: Please update docs/features/compatibility_matrix.md
79
# If the feature combo become valid
80
if not model_config.enforce_eager:
81
model_config.enforce_eager = True
vllm/spec_decode/spec_decode_worker.py
+1
-1
@@ -114,7 +114,7 @@ def create_spec_worker(*args, **kwargs) -> "SpecDecodeWorker":
114
return spec_decode_worker
115
116
117
-# Reminder: Please update docs/source/features/compatibility_matrix.md
117
+# Reminder: Please update docs/features/compatibility_matrix.md
118
# If the feature combo become valid
119
class SpecDecodeWorker(LoRANotSupportedWorkerBase):
120
"""Worker which implements speculative decoding.
vllm/utils.py
+1
-1
@@ -85,7 +85,7 @@ MULTIMODAL_MODEL_MAX_NUM_BATCHED_TOKENS = 5120
85
86
# Exception strings for non-implemented encoder/decoder scenarios
87
88
-# Reminder: Please update docs/source/features/compatibility_matrix.md
88
+# Reminder: Please update docs/features/compatibility_matrix.md
89
# If the feature combo become valid
90
91
STR_NOT_IMPL_ENC_DEC_SWA = \
vllm/worker/multi_step_model_runner.py
+1
-1
@@ -824,7 +824,7 @@ def _pythonize_sampler_output(
824
825
for sgdx, (seq_group,
826
sample_result) in enumerate(zip(seq_groups, samples_list)):
827
- # Reminder: Please update docs/source/features/compatibility_matrix.md
827
+ # Reminder: Please update docs/features/compatibility_matrix.md
828
# If the feature combo become valid
829
# (Check for Guided Decoding)
830
if seq_group.sampling_params.logits_processors:
vllm/worker/utils.py
+1
-1
@@ -14,7 +14,7 @@ def assert_enc_dec_mr_supported_scenario(
14
a supported scenario.
15
'''
16
17
- # Reminder: Please update docs/source/features/compatibility_matrix.md
17
+ # Reminder: Please update docs/features/compatibility_matrix.md
18
# If the feature combo become valid
19
20
if enc_dec_mr.cache_config.enable_prefix_caching: