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 14, 05:30 UTC
[Misc] Add @tdoublep as a maintainer of hybrid model and Triton-attention related code (#23122)
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
Thomas Parnell committed
Aug 19, 2025 at 02:31 UTC
95e30951365d4e5be35117f4fe9c2ffea27975ab
1 file changed
+9
.github/CODEOWNERS
+9
@@ -10,6 +10,7 @@
10
/vllm/worker/worker.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
11
/vllm/model_executor/layers/sampler.py @zhuohan123 @youkaichao @alexm-redhat @comaniac @njhill
12
/vllm/model_executor/layers/quantization @mgoin @robertgshaw2-redhat @tlrmchlsmth @yewentao256
13
+/vllm/model_executor/layers/mamba @tdoublep
14
/vllm/multimodal @DarkLight1337 @ywang96
15
/vllm/vllm_flash_attn @LucasWilkinson
16
/vllm/lora @jeejeelee
@@ -25,6 +26,7 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson
26
# vLLM V1
27
/vllm/v1 @WoosukKwon @robertgshaw2-redhat @njhill @ywang96 @comaniac @alexm-redhat
28
/vllm/v1/structured_output @mgoin @russellb @aarnphm
29
+/vllm/v1/attention/backends/triton_attn.py @tdoublep
30
31
# Test ownership
32
/.buildkite/lm-eval-harness @mgoin @simon-mo
@@ -44,6 +46,7 @@ CMakeLists.txt @tlrmchlsmth @LucasWilkinson
46
/tests/v1/structured_output @mgoin @russellb @aarnphm
47
/tests/weight_loading @mgoin @youkaichao @yewentao256
48
/tests/lora @jeejeelee
49
+/tests/models/language/generation/test_hybrid.py @tdoublep
50
51
# Docs
52
/docs @hmellor
@@ -72,3 +75,9 @@ mkdocs.yaml @hmellor
75
/vllm/model_executor/models/pixtral*.py @patrickvonplaten
76
/vllm/transformers_utils/configs/mistral.py @patrickvonplaten
77
/vllm/transformers_utils/tokenizers/mistral.py @patrickvonplaten
78
+
79
+# Kernels
80
+/vllm/attention/ops/chunked_prefill_paged_decode.py @tdoublep
81
+/vllm/attention/ops/triton_unified_attention.py @tdoublep
82
+
83
+