[CI Failure] Fix test_flashinfer_cutlass_mxfp4_mxfp8_fused_moe (vllm-…

…project#24750)

Signed-off-by: mgoin <mgoin64@gmail.com>

[Compilation Bug] Fix Inductor Graph Output with Shape Issue (vllm-pr…

…oject#24772)

Signed-off-by: yewentao256 <zhyanwentao@126.com>

Merge branch 'main' into wye-refactor-quant-folder

[Bugfix] fixes the causal_conv1d_update kernel update non-speculative…

… decoding cases (vllm-project#24680)

Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>

this is only used to fix nightly wheel version, not a real release ca…

…ndidate

Do not use eval() to convert unknown types (vllm-project#23266)

Signed-off-by: Russell Bryant <rbryant@redhat.com>
Signed-off-by: simon-mo <simon.mo@hey.com>

Use Blackwell FlashInfer MXFP4 MoE by default if available (vllm-proj…

…ect#23008)

Signed-off-by: mgoin <mgoin64@gmail.com>

fix: gptq marlin weight loading failure (vllm-project#23066)

Add think chunk (vllm-project#21333)

Signed-off-by: Julien Denize <julien.denize@mistral.ai>

Add think chunk (vllm-project#21333)

Signed-off-by: Julien Denize <julien.denize@mistral.ai>