[sync] vLLM 2026-07-18 — 3b39fd284..109b736b8#1713
Open
AlpinDale wants to merge 94 commits into
Open
Conversation
… (#48671) Upstream-vLLM: 3b39fd284aa3bcfcad57736cdd5ddb9eb4b746d2 Co-authored-by: aoshen02 <aoshen@inferact.ai>
…#48527) Upstream-vLLM: b6770d7b544febb9636f1750b77cde8904024728 Co-authored-by: djramic <djoramic@amd.com>
Upstream-vLLM: adce0681181e8e258e2028f1b78ad24718f5c169 Co-authored-by: Charlie Fu <charlifu@amd.com>
Upstream-vLLM: e281ac663a7b772f8b988f270521cd1155f12aa3 Co-authored-by: Bugen Zhao <i@bugenzhao.com>
…#48159) Upstream-vLLM: d119beb1b90804f55e5c1d70f1c1961b8a852bd1 Co-authored-by: Giuseppe Grossi <ggrossi@amd.com>
Upstream-vLLM: 9dd2e72828c13bf088403ed789c4672ffef99564 Co-authored-by: Aarushi Jain <142941703+aarushjain29@users.noreply.github.com>
…n ROCm (#48688) Upstream-vLLM: 7aab6e2684894fb1dba17cae117f8e84c5a7bbb3 Co-authored-by: Turner Jabbour <doubleujabbour@gmail.com>
Upstream-vLLM: 313fae3e89daeb7f5183064912933c328bbb6e48 Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
…rom_request() (#48098) Upstream-vLLM: c0302d949740321c145b6b19852706b6f1b43854 Co-authored-by: Mahad Rehman <114791389+mahadrehmann@users.noreply.github.com>
…nforcing different `kernel_block_size` (#48125) Upstream-vLLM: 66b6c684ab5a32dabcd2e69daaa60ff5d5198392 Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@mistral.ai>
…JSON tool-call parsers (#48034) Upstream-vLLM: 4e04bcbce6f0852017d26660dd2d027de9bed7d5 Co-authored-by: Tahsin Tunan <tahsintunan@gmail.com>
…client` test (#47873) Upstream-vLLM: 1b30ae4ca403588f1df37901d3b530d1fe5c48c4 Co-authored-by: Tahsin Tunan <tahsintunan@gmail.com>
… (#48545) Upstream-vLLM: 5811ed6a05158b5d2f953e3b8d3e3082ee36e635 Co-authored-by: Chaojun Zhang <chaojun.zhang@intel.com>
…-DP-rank control port (#47636) Upstream-vLLM: 615834ee584fd6395f1ac4d2257cb29c47895295 Co-authored-by: liranschour <liranschour@users.noreply.github.com>
…Classification (#47991) Upstream-vLLM: 5810e884f11875b46fbae0871754d855d8891105 Co-authored-by: Gopala-Krishna Char <crgkc.r@gmail.com>
…urrent stream (#48526) Upstream-vLLM: 05eed72aec6c05e6d500c7276b47f7652bb37af6 Co-authored-by: peizhang56 <pei.zhang@amd.com>
Upstream-vLLM: 61141ed265bfef41a0ca19e992567ea980919b96 Co-authored-by: Tomasz Zielinski <85164140+tzielinski-habana@users.noreply.github.com>
Upstream-vLLM: 2fa63e0fff41e2f6187a6d3b9bf49b25f9e065f8 Co-authored-by: Xiaohong (Sean) Chen <seachen@redhat.com>
Upstream-vLLM: 6036bf110a9f4d313c38c257ddcf5b8ba4210383 Co-authored-by: Xiaohong (Sean) Chen <seachen@redhat.com>
…layout (#47770) Upstream-vLLM: 1d99f0f42152a1459e4b54804b44fc1f989558bb Co-authored-by: Giuseppe Grossi <ggrossi@amd.com>
…as (#48252) Upstream-vLLM: 43cd340247f442cfccaa8bef24c0867493af8e7b Co-authored-by: Sage <80211083+sagearc@users.noreply.github.com>
Upstream-vLLM: de100ffb622f435b727177fc751c8de5f41c4d3b Co-authored-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
…_model proposer (#47460) Upstream-vLLM: b7950e798f2094b1163c22787c0ba2e3231bf01b Co-authored-by: avalliappan-nvidia <avalliappan@nvidia.com>
…ark) on Blackwell (#48167) Upstream-vLLM: ecf4aa5ce2ccd4069f12318ca9d3fcef7c9f6257 Co-authored-by: Michael Goin <mgoin64@gmail.com>
Upstream-vLLM: 3034c8d389edaf4a131c30304eee76bb9aef5022 Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@mistral.ai>
…r TRT-LLM shuffle (M%128) (#46880) Upstream-vLLM: 2bd8957627bfb5668c46f2bc359bef47d371270c Co-authored-by: Mike G <mgschwind+mikekg@nvidia.com>
…(#47718) Upstream-vLLM: eb33ff34dd6501f23f5be92a5c842d34ab05d7fd Co-authored-by: kliuae <17350011+kliuae@users.noreply.github.com>
…r GLM5.2 (#47881) Upstream-vLLM: 4238b011a76629104b00ecf999d5390b94a1d289 Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
…ormance improvement (#47463) Upstream-vLLM: 2dab187f756f2c4549fd5e57019413f3415365a4 Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
…tion (#47309) Upstream-vLLM: 0becb7486b3e368bb9c0e56b1ca3ccf8ee53911a Co-authored-by: ruikangliu <69446971+ruikangliu@users.noreply.github.com>
… in Quark OCP-MX (#48015) Upstream-vLLM: 7d56fe2adc93d20a01be30eaa275e156ecdd4468 Co-authored-by: music-dino <111048524+music-dino@users.noreply.github.com>
…rams (#47679) Upstream-vLLM: ce653856186e80b2a927a611cb184aeac64d2862 Co-authored-by: Srinivas Krovvidi <194645829+Srinivasoo7@users.noreply.github.com>
Upstream-vLLM: 251f7e478e8eb0c90a01eb8fff40056da2aa3ff7 Co-authored-by: Woosuk Kwon <woosuk@inferact.ai>
Upstream-vLLM: 626c90b2d51e4fdbf4fac361ffa0e52561e40202 Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Upstream-vLLM: f61163e6c736ba2660982769c1d729411b44490e Co-authored-by: Woosuk Kwon <woosuk@inferact.ai>
Upstream-vLLM: 02bf9c7907b491146e2a79c3839af243c4478e00 Co-authored-by: xiao-llm <xiao.yu.dc@outlook.com>
…48868) Upstream-vLLM: efa2e424f659dd6bbc07982bc155f265e32532cd Co-authored-by: Shangdi Yu <yushangdi123@gmail.com>
…ressure instead of failing the read (#47495) Upstream-vLLM: 971dac2caa827cb34dc682791ac6e15f6dbdab84 Co-authored-by: limeward <32970461+edwinlim0919@users.noreply.github.com>
Upstream-vLLM: fb5ec0dc9edfdba54882575e33e2a4215bd295b9 Co-authored-by: Woosuk Kwon <woosuk@inferact.ai>
…Int (#48785) Upstream-vLLM: ab3c1aedf3b9677c1e43b391478554a4b110419e Co-authored-by: HDCharles <39544797+HDCharles@users.noreply.github.com>
Upstream-vLLM: c95c6630499f96ddee97a87ba1ca4e6b85afe0fe Co-authored-by: Michael Goin <mgoin64@gmail.com>
Upstream-vLLM: 4a394bfcda817562af9a9e070e7c50ae57143b99 Co-authored-by: Hangrui Cao <50705298+DiegoCao@users.noreply.github.com>
…+ KV Cache runtime fusion for Qwen3-30B-A3B on ROCM_AITER_FA, and ROCM_AITER_UNIFIED_ATTN (#42749) Upstream-vLLM: 2cab53ddeea3851a083beff8c192f068c4518b67 Co-authored-by: Jack Hu <jack.hu@amd.com>
Upstream-vLLM: f17be06fbe42bab61522868a78d044fd0d2e3547 Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Upstream-vLLM: 67f9046e4a12d939ffad0abc917585a00ef5b42d Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
…#44749) Upstream-vLLM: b88abb5036aa6b7191a3e9341ab652e6562a49d2 Co-authored-by: Daoyuan Li <94409450+DaoyuanLi2816@users.noreply.github.com>
Upstream-vLLM: fe784ff22e630a31fd798f392b01e0a75c18f047 Co-authored-by: Thien Tran <gau.nernst@yahoo.com.sg>
Upstream-vLLM: f3e9497e921a16741401c5e93af0c2c29ea74907 Co-authored-by: Woosuk Kwon <woosuk@inferact.ai>
Upstream-vLLM: ee8f36d0b3dc36ad1751c2193b30feb2ba11e0da Co-authored-by: Woosuk Kwon <woosuk@inferact.ai>
… reasoning to finish (#48873) Upstream-vLLM: 67fe73b2b47e2927af1e5ec6fb91e16e3b075940 Co-authored-by: Micah Williamson <micah.williamson@amd.com>
…617) Upstream-vLLM: 4d4e04f45221709079316599f01b33a5233deaf2 Co-authored-by: Martin Hickey <martin.hickey@ie.ibm.com>
…orch.compile (#48901) Upstream-vLLM: 3b6c96a101651c78737b23b2bd8f1026c02f22e2 Co-authored-by: Sebastian Woo <43553500+woosebastian@users.noreply.github.com>
…t heterogeneous KV cache groups (#48878) Upstream-vLLM: 472d330c214f88c1a2ba4ade75d21f25782e8bf5 Co-authored-by: Debasish Mohanty <139056705+Debasish-87@users.noreply.github.com>
…eloads (#48251) Upstream-vLLM: 17fdd421009092a569e9ed22346b932e55824fb5 Co-authored-by: aoshen02 <aoshen@inferact.ai>
Upstream-vLLM: 9354f222042986addf20709e5274fc26e0d09745 Co-authored-by: Roy Wang <jasonailu87@gmail.com>
Upstream-vLLM: fb1d8ccaf5bc46935031ef074950eaa8232cd15e Co-authored-by: Aaron Hao <ahao@anyscale.com>
Upstream-vLLM: 26c909ed74a6298952d0c3191fbfdf2b513d9e1d Co-authored-by: Zhang Jian <jianmusings@gmail.com>
…48281) Upstream-vLLM: 426d48bfa149582664d48f89df21ec9beae5c37b Co-authored-by: Chang Guo <changg@nvidia.com>
…xed image/video inputs (#46213) Upstream-vLLM: 69d4f5ef6323a44621b10e0f269c2775db510cc5 Co-authored-by: wendadawen <130649302+wendadawen@users.noreply.github.com>
Upstream-vLLM: 109b736b8681dff6e2f547d89362afc3a9c77c00
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Automated upstream sync: 3b39fd284aa3..109b736b8681