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 7, 15:15 UTC
[Bug] Fix outdated links in source code (#35314)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Wentao Ye committed
Feb 26, 2026 at 22:50 UTC
062b7896324df60d5c03f66201902122f6cb8901
5 files changed
+11
-7
.github/mergify.yml
+1
-2
@@ -259,8 +259,7 @@ pull_request_rules:
259
- files=benchmarks/run_structured_output_benchmark.sh
260
- files=docs/features/structured_outputs.md
261
- files=examples/offline_inference/structured_outputs.py
262
- - files=examples/online_serving/openai_chat_completion_structured_outputs.py
263
- - files=examples/online_serving/openai_chat_completion_structured_outputs_with_reasoning.py
262
+ - files=examples/online_serving/structured_outputs/structured_outputs.py
263
- files~=^tests/v1/structured_output/
264
- files=tests/v1/entrypoints/llm/test_struct_output_generate.py
265
- files~=^vllm/v1/structured_output/
docs/design/metrics.md
+1
-1
@@ -656,7 +656,7 @@ vLLM has support for OpenTelemetry tracing:
656
- Added by <https://github.com/vllm-project/vllm/pull/4687> and reinstated by <https://github.com/vllm-project/vllm/pull/20372>
657
- Configured with `--oltp-traces-endpoint` and `--collect-detailed-traces`
658
- [OpenTelemetry blog post](https://opentelemetry.io/blog/2024/llm-observability/)
659
-- [User-facing docs](../examples/online_serving/opentelemetry.md)
659
+- [User-facing docs](../../examples/online_serving/opentelemetry/README.md)
660
- [Blog post](https://medium.com/@ronen.schaffer/follow-the-trail-supercharging-vllm-with-opentelemetry-distributed-tracing-aa655229b46f)
661
- [IBM product docs](https://www.ibm.com/docs/en/instana-observability/current?topic=mgaa-monitoring-large-language-models-llms-vllm-public-preview)
662
tools/profiler/print_layerwise_table.py
+4
-1
@@ -33,7 +33,10 @@ if __name__ == "__main__":
33
"--json-trace",
34
type=str,
35
required=True,
36
- help="json trace file output by examples/offline_inference/profiling.py",
36
+ help=(
37
+ "JSON trace file generated by scripts that use "
38
+ "vllm.profiler.layerwise_profile"
39
+ ),
40
)
41
parser.add_argument(
42
"--phase",
tools/profiler/visualize_layerwise_profile.py
+4
-2
@@ -564,8 +564,10 @@ if __name__ == "__main__":
564
"--json-trace",
565
type=str,
566
required=True,
567
- help="json trace file output by \
568
- examples/offline_inference/profiling.py",
567
+ help=(
568
+ "JSON trace file generated by scripts that use "
569
+ "vllm.profiler.layerwise_profile"
570
+ ),
571
)
572
parser.add_argument(
573
"--output-directory", type=str, required=False, help="Directory to output plots"
vllm/model_executor/models/config.py
+1
-1
@@ -213,7 +213,7 @@ class NomicBertModelConfig(VerifyAndUpdateConfig):
213
"Nomic context extension is disabled. "
214
"Changing max_model_len from %s to %s. "
215
"To enable context extension, see: "
216
- "https://github.com/vllm-project/vllm/tree/main/examples/offline_inference/context_extension.html",
216
+ "https://github.com/vllm-project/vllm/tree/main/examples/offline_inference/context_extension.py",
217
max_model_len_before,
218
model_config.max_model_len,
219
)