From e390742c5906df0edad8fc77b203e0623559ce79 Mon Sep 17 00:00:00 2001 From: XueLiang Yang <102161631+xueliangyang-oeuler@users.noreply.github.com> Date: Thu, 19 Mar 2026 20:05:07 +0800 Subject: [PATCH] =?UTF-8?q?Fix=20KV=20Offloading=20+=20MLA=20AssertionErro?= =?UTF-8?q?r=20by=20using=20num=5Fkv=5Fheads=3D1=20in=20cpu=E2=80=A6=20(#3?= =?UTF-8?q?7536)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: xueliangyang-oeuler Co-authored-by: xueliangyang-oeuler --- vllm/v1/kv_offload/worker/cpu_gpu.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/v1/kv_offload/worker/cpu_gpu.py b/vllm/v1/kv_offload/worker/cpu_gpu.py index 4ce3574371..69a827a870 100644 --- a/vllm/v1/kv_offload/worker/cpu_gpu.py +++ b/vllm/v1/kv_offload/worker/cpu_gpu.py @@ -240,7 +240,7 @@ class CpuGpuOffloadingHandlers: gpu_shape = gpu_tensor.shape attn_backend = attn_backends[layer_name] test_shape = attn_backend.get_kv_cache_shape( - num_blocks=1234, block_size=16, num_kv_heads=8, head_size=256 + num_blocks=1234, block_size=16, num_kv_heads=1, head_size=256 ) has_layers_dim = False