vllm-project/vllm

Pull Requests Commits

Fix Python version regex quoting in Dockerfile

khluu committed 107 days ago

7c36acf7

Build Python from source instead of using deadsnakes PPA

khluu committed 107 days ago

9f876610

khluu committed 107 days ago

e7c93594

khluu committed 107 days ago

747e3795

[DSV4] Add knob to enable pre-attn gemm (#41443)

zyongye committed 107 days ago

513b5f82

khluu committed 107 days ago

135722d5

[ROCm][Bugfix][GPTOSS]: fix input_ids and expert_map args for quark w4a8 gptoss (#41165)

Rohan138 committed 107 days ago

7758daca

[Perf] Intergrate Tile Kernels `head_compute_mix_kernel` for Deepseek-V4 (#41255)

Isotr0py committed 107 days ago

5f79634b

[Bugfix] Pass reasoning parser kwargs to structured output (#41199)

BugenZhao committed 107 days ago

05adde99

[Bugfix] BailingMoeV2.5: rotate full qk_rope_head_dim in MLA RoPE (#41185)

ZJY0516 committed 107 days ago

9cdcda5f

[Bugfix ] fix bailing_moe_linear (#40859)

ghphotoframe committed 107 days ago

a5d4a264

zyongye committed 108 days ago

a5e0ea09

[DSV4] Add BF16 and MXFP8 A2A support for flashinfer a2a one sided (#40960)

zyongye committed 108 days ago

28a1a85d

[Bugfix] Fix persistent_topk cooperative deadlock at TopK=1024 (#41189)

zyongye committed 108 days ago

82ccd9f3

[DSv4] Use `cvt` PTX for FP32->FP4 conversion (#41015)

gau-nernst committed 108 days ago

d9f3961a

[Bugfix] Fix repeated DSv4 RoPE cache initialization (#41148)

jeejeelee committed 108 days ago

4fee1390

[Bugfix] DSV32/V4 add missing type conversion for non-streaming tool calls (#41198)

chaunceyjiang committed 108 days ago

bf590b6a

[Bugfix] Fix max_num_batched_token not captured in cuda graph (#40734)

wzhao18 committed 108 days ago

66df2c03

[Bugfix] fix inductor error for dpsk v4 (#41135)

ZJY0516 committed 108 days ago

17896d81

[Bugfix] Fix Deepseek V4 import error due to AOT compile cache loading (#41090)

wzhao18 committed 108 days ago

1a131a06

[DSV4] Align aux stream API with DeepseekV4DecoderLayer (#41171)

zixi-qi committed 108 days ago

6fffb0b8

[Core] Account for `num_gpu_blocks_override` in `max_model_len` checks (#41069)

njhill committed 108 days ago

a076bd1e

[DSV4] Enable Multi-stream for Pre-Attn GEMM (#41061)

zyongye committed 108 days ago

c244b6fc

[Model][DSV4] Support base model (#41006)

jeejeelee committed 108 days ago

15cc87d2

Auto-disable expandable_segments around cumem memory pool (#40812)

youkaichao committed 108 days ago

c68fa39c

Temporary disable persistent topk (#41442)

zyongye committed 108 days ago

Verified 05ebca52

[Docker] Install numactl CLI in CUDA runtime image (#41032)

zhewenl committed 111 days ago

88d34c64

[DSV4] Add silu clamp limit to shared expert (#40950)

zyongye committed 112 days ago

b8160878

[Bugfix] Cap SWA/chunked-local runtime admission to startup pool-sizing bound (#40946)

Dao007forever committed 112 days ago

84c276d7

Revert "[Frontend] Remove frontend pooling multi task support. (#37861)"

khluu committed 112 days ago

5eb36575

Older