Add ovis models support with default buckets#846

Open

testdig wants to merge 23 commits intovllm-project:mainfrom

Commits on Jan 21, 2026

Commits on Jan 29, 2026

Add copyright header
testdig
committed
Added Qwen3 Test (#736)

authored and
testdig
committed
Interleaved sliding window fix (#805)

rsmyrek
authored and
testdig
committed
[GAUDISW-245665] fix diverge from vllm in multiModalBudget (#837)

authored and
testdig
committed
KV cache sharing for HPU (#834)

authored and
testdig
committed
Fix for #32077 (#851)

authored and
testdig
committed
Resolve qwen25 vl accuracy regression (#831)

tvoas
authored and
testdig
committed
DP: Fix for torch.compile (#722)

authored and
testdig
committed
Remove unused test utils (#864)

microslaw
authored and
testdig
committed
Add support for chunked attention (#821)

kfojcik-intel
authored and
testdig
committed
Fix Llama4 shape mismatch for 32k+ context window (#842) (#855)

afierka-intel
authored and
testdig
committed
Fix HPU model runner profile_run to work with dynamic kv-cache scales (#852)

authored and
testdig
committed
Revert "skip HPU graphs for long prefills" (#850)

authored and
testdig
committed
Implement bucket corrector for Mamba chunk size (#886)

jbyczkow
authored and
testdig
committed
Draft: Add FlashAttention online merge in Unified Attention (#785)

authored and
testdig
committed
Enable support for prefill side kv_layout and block_size update (#867)

yeonsily
authored and
testdig
committed
Merge branch 'main' into ovis
testdig
authored
Merge branch 'main' into ovis
testdig
authored

Commits on Feb 2, 2026

Commits on Feb 4, 2026

Merge branch 'main' into ovis
michalkuligowski
authored