Merged
Conversation
orangetin
referenced
this pull request
in togethercomputer/vllm-ttgi
Sep 14, 2023
add rope scaling as a cli arg so openai server can load rope scaled models
xiangyuT
pushed a commit
to xiangyuT/vllm
that referenced
this pull request
Oct 18, 2023
l1cacheDell
pushed a commit
to CaspianFang/vllm
that referenced
this pull request
Nov 15, 2023
hongxiayang
referenced
this pull request
in hongxiayang/vllm
Feb 13, 2024
ilya-lavrenov
referenced
this pull request
in ilya-lavrenov/vllm
Feb 19, 2024
Deterministic OpenVINO inference
daniel-geon-park
added a commit
to gmlwns2000/vllm-timber
that referenced
this pull request
Apr 15, 2024
mzusman
pushed a commit
to mzusman/vllm
that referenced
this pull request
Apr 16, 2024
BA-78554: Jurassic 2.5 * worked on jurasic2.5 configuration file, updated jurassic2_5 modeling file to support alternating experts/attn layers * finished working the forward pass of jurassic3.py * finished working the forward pass of jurassic3.py * finished working the forward pass of jurassic3.py * jurassic_3 modeling file works, uses dummy weights initialized by "dummy" flag. Tokenizer raises issues, for now copying the mixtral tokenizer * changed default tokenizer vocab values, loading of custom .pt weight files works. * removed notebook * merging master to jurassic-2.5 to reset head * Merge branch 'master' into jurassic-2.5 * align to master Approved-by: Tomer Asida Approved-by: Mor Zusman
Bellk17
added a commit
to Bellk17/vllm
that referenced
this pull request
May 10, 2024
Triton compilation fix
This was referenced Jul 5, 2024
1 task
1 task
1 task
saifmb0
added a commit
to saifmb0/vllm
that referenced
this pull request
Mar 29, 2026
…t#38472) Add 17 unit tests covering the KV cache memory-management fixes from the previous commit. Tests run without NCCL, ZMQ, or CUDA by constructing a minimal engine object via object.__new__ and stubbing the two compiled extensions (vllm._C, vllm._C_stable_libtorch) in sys.modules. Test classes ------------ TestRecvTensorPopsFromStore (4 tests) - entry_absent_after_recv_tensor (regression guard vllm-project#1) - buffer_size_decremented_after_recv_tensor - multiple_layers_all_popped - recv_store_does_not_grow_under_sustained_load (high-QPS OOM scenario) TestRecvTensorPoolBackedEntries (4 tests) - pool_free_called_immediately (regression guard vllm-project#2) - pool_entry_removed_from_recv_store - pool_free_not_called_for_none_tensor (edge case: listener OOM path) - pool_free_called_once_per_entry_not_twice TestGetFinishedStragglerCleanup (6 tests) - straggler_gpu_tensor_removed_and_buffer_decremented - straggler_pool_tensor_freed - multiple_straggler_layers_all_cleaned - only_finished_request_cleaned_not_others - noop_when_tensors_already_consumed - send/recv_tracking_dict_cleaned_by_get_finished TestGetFinishedNoCompileLayersParamUnused (1 test) - Verifies straggler cleanup works even with no_compile_layers={} (the old iteration approach would silently skip all cleanup here) TestPoolFreeCalledOnce (1 test) - combined_lifecycle_exactly_one_free: ensures pool.free is called exactly once across recv_tensor + get_finished, not twice. Reproduction of original bugs confirmed: - Old recv_tensor (read without pop) leaves entry in recv_store: CONFIRMED - Old recv_tensor (no pool.free) leaks pinned RAM: CONFIRMED All 17 tests pass against the fixed code. Co-authored-by: GitHub Copilot Signed-off-by: saif <contact@saifmb.com>
Damon-Salvetore
pushed a commit
to Damon-Salvetore/vllm
that referenced
this pull request
Mar 31, 2026
…n-files-overview Add Chinese documentation for vLLM framework structure and GEMM call chains
anand-nv
referenced
this pull request
in anand-nv/vllm
Apr 1, 2026
[WIP] Fix Mamba state contamination in KV cache block reuse
This was referenced Apr 2, 2026
7 tasks
1 task
1 task
jefp
added a commit
to jefp/vllm
that referenced
this pull request
Apr 9, 2026
Rebase of PR vllm-project#33315 onto current main. Adds max_tokens_per_doc parameter to rerank requests, matching Cohere and Jina rerank APIs. Documents longer than this limit are truncated before scoring. Handles all three cross-encoder code paths: - Cross-encoder with sep token (tokenizer built-in truncation) - Chat template / Jinja path (text truncation before template) - Score template path (text truncation before template) Also supports offline usage via PoolingParams(extra_kwargs={"max_tokens_per_doc": N}). Addresses reviewer feedback from original PR: - Offline support via PoolingParams (noooop vllm-project#1) - Score template compatibility tests (noooop vllm-project#2) - Tests across BAAI/bge-reranker-base, BAAI/bge-reranker-v2-gemma, and Qwen/Qwen3-Reranker-0.6B Original PR: vllm-project#33315 Original author: hustxiayang Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Jesus Federico <jefp@amazon.com>
1 task
jefp
added a commit
to jefp/vllm
that referenced
this pull request
Apr 10, 2026
Rebase of PR vllm-project#33315 onto current main. Adds max_tokens_per_doc parameter to rerank requests, matching Cohere and Jina rerank APIs. Documents longer than this limit are truncated before scoring. Handles all three cross-encoder code paths: - Cross-encoder with sep token (tokenizer built-in truncation) - Chat template / Jinja path (text truncation before template) - Score template path (text truncation before template) Also supports offline usage via PoolingParams(extra_kwargs={"max_tokens_per_doc": N}). Addresses reviewer feedback from original PR: - Offline support via PoolingParams (noooop vllm-project#1) - Score template compatibility tests (noooop vllm-project#2) - Tests across BAAI/bge-reranker-base, BAAI/bge-reranker-v2-gemma, and Qwen/Qwen3-Reranker-0.6B Original PR: vllm-project#33315 Original author: hustxiayang Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Signed-off-by: Jesus Federico <jefp@amazon.com>
6 tasks
4 tasks
1 task
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
1 participant
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.
This PR fixes a bug in supporting OPT-350m/OPT-6.7b/OPT-13b and OPT-IML models.
The bug happened because our model code didn't include some methods that were required to tie the input and output embeddings.