refactor: 将 kv_compression 的 Triton 内核迁移到 vllm/v1/kv_compression

9db5ff3b · laibao · 3bc7eb74 · 3bc7eb74 · 9db5ff3b · 9db5ff3b
Commit 9db5ff3b authored Jan 27, 2026 by laibao
3 changed files
--- a/vllm/v1/attention/kv_compression/__init__.py
+++ b/vllm/v1/attention/kv_compression/__init__.py
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
-
--- a/vllm/v1/attention/kv_compression/kv_cache_triton.py
+++ b/vllm/v1/attention/kv_compression/kv_cache_triton.py
--- a/vllm/v1/attention/kv_compression/snapkv_triton.py
+++ b/vllm/v1/attention/kv_compression/snapkv_triton.py