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, 03:30 UTC
[Misc] Add rules to label Speculative Decoding Related PRs (#20406)
Signed-off-by: Lifan Shen <lifans@meta.com>
Lifans committed
Jul 2, 2025 at 23:56 UTC
220aee902a291209f2975d4cd02dadcc6749ffe6
1 file changed
+6
.github/mergify.yml
+6
@@ -156,8 +156,14 @@ pull_request_rules:
156
conditions:
157
- or:
158
- files~=^vllm/spec_decode/
159
+ - files~=^vllm/v1/spec_decode/
160
- files=vllm/model_executor/layers/spec_decode_base_sampler.py
161
- files~=^tests/spec_decode/
162
+ - files~=^tests/v1/spec_decode/
163
+ - files~=^examples/.*(spec_decode|mlpspeculator|eagle|speculation).*\.py
164
+ - files~=^vllm/model_executor/models/.*eagle.*\.py
165
+ - files=vllm/model_executor/models/mlp_speculator.py
166
+ - files~=^vllm/transformers_utils/configs/(eagle|medusa|mlp_speculator)\.py
167
actions:
168
label:
169
add: