docs(governance): refresh committers list, add TSC note, update project leads (#51300)
Simon Mo committed
Aug 6, 2026 at 16:28 UTC
9bca7d840d7fa4677e58e7d163ddd191cccc40b7
2 files changed
+40
-34
docs/governance/committers.md
+33
-30
@@ -9,71 +9,81 @@ You can also refer to the [CODEOWNERS](https://github.com/vllm-project/vllm/blob
9
We try to summarize each committer's role in vLLM in a few words. In general, vLLM committers cover a wide range of areas and help each other in the maintenance process.
10
Please refer to the later section about Area Owners for exact component ownership details.
11
Sorted alphabetically by GitHub handle:
12
-
13
-- [@22quinn](https://github.com/22quinn): RL API
12
- [@aarnphm](https://github.com/aarnphm): Structured output
15
-- [@alexm-redhat](https://github.com/alexm-redhat): Performance
13
+- [@AndreasKaratzas](https://github.com/AndreasKaratzas): ROCm / AMD GPU integration, CI
14
- [@ApostaC](https://github.com/ApostaC): Connectors, offloading
15
- [@bbrowning](https://github.com/bbrowning): Tool use and reasoning parser
16
- [@benchislett](https://github.com/benchislett): Engine core and spec decode
17
- [@bigPYJ1151](https://github.com/bigPYJ1151): Intel CPU/XPU integration
18
+- [@BoyuanFeng](https://github.com/BoyuanFeng): Compilation, CUDAGraph
19
- [@BugenZhao](https://github.com/BugenZhao): Rust frontend
20
- [@chaunceyjiang](https://github.com/chaunceyjiang): Tool use and reasoning parser
21
- [@DarkLight1337](https://github.com/DarkLight1337): Multimodality, API server
23
-- [@esmeetu](https://github.com/esmeetu): developer marketing, community
22
- [@dllehr-amd](https://github.com/dllehr-amd): AMD integration
23
+- [@esmeetu](https://github.com/esmeetu): developer marketing, community
24
+- [@gshtras](https://github.com/gshtras): ROCm / AMD GPU integration
25
+- [@Harry-Chen](https://github.com/Harry-Chen): Build, CI, CUDA architecture
26
- [@heheda12345](https://github.com/heheda12345): Hybrid memory allocator
27
- [@hmellor](https://github.com/hmellor): Hugging Face integration, documentation
28
+- [@hongxiayang](https://github.com/hongxiayang): AMD GPU / ROCm integration
29
- [@houseroad](https://github.com/houseroad): Engine core and Llama models
30
- [@Isotr0py](https://github.com/Isotr0py): Multimodality, new model support
31
+- [@ivanium](https://github.com/ivanium): KV Connector and offload
32
- [@jeejeelee](https://github.com/jeejeelee): LoRA, new model support
33
- [@jikunshang](https://github.com/jikunshang): Intel CPU/XPU integration
34
- [@khluu](https://github.com/khluu): CI infrastructure
32
-- [@KuntaiDu](https://github.com/KuntaiDu): KV Connector
35
- [@LucasWilkinson](https://github.com/LucasWilkinson): Kernels and performance
34
-- [@luccafong](https://github.com/luccafong): Llama models, speculative decoding, distributed
36
- [@markmc](https://github.com/markmc): Observability
37
- [@MatthewBonanni](https://github.com/MatthewBonanni): Kernels and performance
38
- [@mgoin](https://github.com/mgoin): Quantization and performance
39
- [@NickLucche](https://github.com/NickLucche): KV connector
40
- [@njhill](https://github.com/njhill): Distributed, API server, engine core
41
- [@noooop](https://github.com/noooop): Pooling models
41
-- [@patrickvonplaten](https://github.com/patrickvonplaten): Mistral models, new model support
42
+- [@orozery](https://github.com/orozery): KV Connector and offload
43
- [@pavanimajety](https://github.com/pavanimajety): NVIDIA GPU integration
44
- [@ProExpertProg](https://github.com/ProExpertProg): Compilation, startup UX
45
- [@robertgshaw2-redhat](https://github.com/robertgshaw2-redhat): Core, distributed, disagg
45
-- [@ruisearch42](https://github.com/ruisearch42): Pipeline parallelism, Ray Support
46
- [@russellb](https://github.com/russellb): Structured output, engine core, security
47
- [@sfeng33](https://github.com/sfeng33): Tool use and reasoning parser
48
-- [@sighingnow](https://github.com/sighingnow): Qwen models, new model support
48
+- [@shen-shanshan](https://github.com/shen-shanshan): AMD GPU / ROCm integration, multimodality, ViT CUDA graph
49
- [@simon-mo](https://github.com/simon-mo): Project lead, API entrypoints, community
50
- [@tdoublep](https://github.com/tdoublep): State space models
51
- [@tjtanaa](https://github.com/tjtanaa): AMD GPU integration
52
- [@tlrmchlsmth](https://github.com/tlrmchlsmth): Kernels and performance, distributed, disagg
53
+- [@tomeras91](https://github.com/tomeras91): MoE, Mamba, Ray
54
+- [@vadiklyutiy](https://github.com/vadiklyutiy): CI, performance, kernels
55
- [@WoosukKwon](https://github.com/WoosukKwon): Project lead, engine core
56
+- [@xinli-sw](https://github.com/xinli-sw): Quantization and model support
57
+- [@xuechendi](https://github.com/xuechendi): Intel CPU/XPU integration, KV connector
58
- [@yaochengji](https://github.com/yaochengji): TPU integration
55
-- [@yeqcharlotte](https://github.com/yeqcharlotte): Benchmark, Llama models
59
- [@yewentao256](https://github.com/yewentao256): Kernels and performance
60
- [@Yikun](https://github.com/Yikun): Pluggable hardware interface
61
- [@youkaichao](https://github.com/youkaichao): Project lead, distributed, compile, community
62
- [@ywang96](https://github.com/ywang96): Multimodality, benchmarks
63
- [@zhuohan123](https://github.com/zhuohan123): Project lead, RL integration, numerics
64
+- [@ZJY0516](https://github.com/ZJY0516): State space models, spec decode, new model support
65
- [@zou3519](https://github.com/zou3519): Compilation
62
-- [@BoyuanFeng](https://github.com/BoyuanFeng): Compilation, CUDAGraph
63
-- [@xuechendi](https://github.com/xuechendi): Intel CPU/XPU integration, KV connector
66
+- [@zyongye](https://github.com/zyongye): MoE kernels and quantization
67
68
### Emeritus Committers
69
70
Committers who have contributed to vLLM significantly in the past (thank you!) but no longer active:
68
-
71
+- [@22quinn](https://github.com/22quinn): RL API
72
+- [@alexm-redhat](https://github.com/alexm-redhat): Performance
73
- [@andoorve](https://github.com/andoorve): Pipeline parallelism
74
- [@cadedaniel](https://github.com/cadedaniel): Speculative decoding
75
- [@comaniac](https://github.com/comaniac): KV cache management, pipeline parallelism
76
+- [@KuntaiDu](https://github.com/KuntaiDu): KV Connector
77
- [@LiuXiaoxuanPKU](https://github.com/LiuXiaoxuanPKU): Speculative decoding
78
+- [@luccafong](https://github.com/luccafong): Llama models, speculative decoding, distributed
79
+- [@patrickvonplaten](https://github.com/patrickvonplaten): Mistral models, new model support
80
- [@pcmoritz](https://github.com/pcmoritz): MoE
81
- [@rkooo567](https://github.com/rkooo567): Chunked prefill
82
+- [@ruisearch42](https://github.com/ruisearch42): Pipeline parallelism, Ray Support
83
+- [@sighingnow](https://github.com/sighingnow): Qwen models, new model support
84
- [@sroy745](https://github.com/sroy745): Speculative decoding
85
- [@Yard1](https://github.com/Yard1): kernels and performance
86
+- [@yeqcharlotte](https://github.com/yeqcharlotte): Benchmark, Llama models
87
- [@zhisbug](https://github.com/zhisbug): Arctic models, distributed
88
89
## Area Owners
@@ -92,9 +102,9 @@ If you have PRs touching the area, please feel free to ping the area owner for r
102
- ModelRunner, Executor, Worker: the abstractions for engine wrapping model implementation
103
- @WoosukKwon, @tlrmchlsmth, @heheda12345, @LucasWilkinson, @ProExpertProg, @MatthewBonanni
104
- KV Connector: Connector interface and implementation for KV cache offload and transfer
95
- - @robertgshaw2-redhat, @njhill, @KuntaiDu, @NickLucche, @ApostaC
105
+ - @robertgshaw2-redhat, @njhill, @NickLucche, @ApostaC, @ivanium, @orozery
106
- Distributed, Parallelism, Process Management: Process launchers managing each worker, and assign them to the right DP/TP/PP/EP ranks
97
- - @youkaichao, @njhill, @WoosukKwon, @ruisearch42
107
+ - @youkaichao, @njhill, @WoosukKwon
108
- Collectives: the usage of nccl and other communication libraries/kernels
109
- @tlrmchlsmth, @youkaichao
110
- Multimodality engine and memory management: core scheduling and memory management concerning vision, audio, and video inputs.
@@ -105,7 +115,7 @@ If you have PRs touching the area, please feel free to ping the area owner for r
115
- Model Interface: The `nn.Module` interface and implementation for various models
116
- @zhuohan123, @mgoin, @simon-mo, @houseroad, @ywang96 (multimodality), @jeejeelee (lora)
117
- Logits Processors / Sampler: The provided sampler class and pluggable logits processors
108
- - @njhill, @houseroad, @22quinn
118
+ - @njhill, @houseroad
119
- Custom Layers: Utility layers in vLLM such as rotary embedding and rms norms
120
- @ProExpertProg
121
- Attention: Attention interface for paged attention
@@ -113,7 +123,7 @@ If you have PRs touching the area, please feel free to ping the area owner for r
123
- FusedMoE: FusedMoE kernel, Modular kernel framework, EPLB
124
- @tlrmchlsmth
125
- Quantization: Various quantization config, weight loading, and kernel.
116
- - @mgoin, @Isotr0py, @yewentao256
126
+ - @mgoin, @Isotr0py, @yewentao256, @zyongye, @xinli-sw
127
- Custom quantized GEMM kernels (cutlass_scaled_mm, marlin, machete)
128
- @tlrmchlsmth, @LucasWilkinson
129
- Multi-modal Input Processing: Components that load and process image/video/audio data into feature tensors
@@ -121,7 +131,7 @@ If you have PRs touching the area, please feel free to ping the area owner for r
131
- torch compile: The torch.compile integration in vLLM, custom passes & transformations
132
- @ProExpertProg, @zou3519, @youkaichao, @BoyuanFeng
133
- State space models: The state space models implementation in vLLM
124
- - @tdoublep, @tlrmchlsmth
134
+ - @tdoublep, @tlrmchlsmth, @ZJY0516, @tomeras91
135
- Reasoning and tool calling parsers
136
- @chaunceyjiang, @aarnphm, @sfeng33, @bbrowning
137
@@ -139,11 +149,11 @@ If you have PRs touching the area, please feel free to ping the area owner for r
149
### Features
150
151
- Spec Decode: Covers model definition, attention, sampler, and scheduler related to n-grams, EAGLE, and MTP.
142
- - @WoosukKwon, @benchislett, @luccafong, @MatthewBonanni
152
+ - @WoosukKwon, @benchislett, @MatthewBonanni
153
- Structured Output: The structured output implementation
154
- @russellb, @aarnphm
155
- RL: The RL related features such as collective rpc, sleep mode, etc.
146
- - @youkaichao, @zhuohan123, @22quinn
156
+ - @youkaichao, @zhuohan123
157
- LoRA: @jeejeelee
158
- Observability: Metrics and Logging
159
- @markmc, @robertgshaw2-redhat, @simon-mo
@@ -154,7 +164,7 @@ If you have PRs touching the area, please feel free to ping the area owner for r
164
- @hmellor
165
- Documentation: @hmellor, @DarkLight1337, @simon-mo
166
- Benchmarks: @ywang96, @simon-mo
157
-- CI, Build, Release Process: @khluu, @njhill, @simon-mo
167
+- CI, Build, Release Process: @khluu, @njhill, @simon-mo, @Harry-Chen, @vadiklyutiy
168
- Security: @russellb
169
170
### External Kernels Integration
@@ -167,21 +177,14 @@ If you have PRs touching the area, please feel free to ping the area owner for r
177
### Integrations
178
179
- Hugging Face: @hmellor, @Isotr0py
170
-- Ray: @ruisearch42
180
+- Ray: (unowned)
181
- NIXL: @robertgshaw2-redhat, @NickLucche
182
173
-### Collaboration with Model Vendors
174
-
175
-- gpt-oss: @heheda12345, @simon-mo, @zhuohan123
176
-- Llama: @luccafong
177
-- Qwen: @sighingnow
178
-- Mistral: @patrickvonplaten
179
-
183
### Hardware
184
185
- Plugin Interface: @youkaichao, @Yikun
186
- NVIDIA GPU: @pavanimajety
184
-- AMD GPU: @gshtras, @tjtanaa
187
+- AMD GPU: @gshtras, @tjtanaa, @hongxiayang, @shen-shanshan, @AndreasKaratzas
188
- Intel CPU/GPU: @jikunshang, @bigPYJ1151, @xuechendi
189
- Google TPU: @yaochengji
190
docs/governance/process.md
+7
-4
@@ -25,7 +25,11 @@ Maintainers form a hierarchy based on sustained, high-quality contributions and
25
26
### Core Maintainers
27
28
-Core Maintainers function like a project planning and decision making committee. In other convention, they might be called a Technical Steering Committee (TSC). In vLLM vocabulary, they are often known as "Project Leads". They meet weekly to coordinate roadmap priorities and allocate engineering resources.
28
+Core Maintainers function like a project planning and decision making committee.
29
+In other convention, they might be called a Technical Steering Committee (TSC).
30
+In vLLM vocabulary, they are often known as "Project Leads".
31
+This committee is also the Technical Steering Committee as defined by Linux Foundation Project Governance.
32
+They meet weekly to coordinate roadmap priorities and allocate engineering resources.
33
34
**Project Leads:**
35
@@ -38,9 +42,8 @@ Core Maintainers function like a project planning and decision making committee.
42
- Michael Goin ([@mgoin](https://github.com/mgoin))
43
- Nick Hill ([@njhill](https://github.com/njhill))
44
- Roger Wang ([@ywang96](https://github.com/ywang96))
41
-- Lu Fang ([@houseroad](https://github.com/houseroad))
42
-- Ye (Charlotte) Qi ([@yeqcharlotte](https://github.com/yeqcharlotte))
43
-- Yihua Cheng ([@ApostaC](https://github.com/ApostaC))
45
+- Lucas Wilkinson ([@LucasWilkinson](https://github.com/LucasWilkinson))
46
+- Xin Li ([@xinli-sw](https://github.com/xinli-sw))
47
48
**Responsibilities:**
49