[doc] add version

CHANGELOG.md

@@ -0,0 +1,30 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [1.0.0] - 2025-09-11
+
+### Added
+- C++ radix tree for fast match, need set "index_accel": true in cache_config
+- sync kernel launch
+- a huge change that move cache engine to a library for accelerator(vLLM e.g.) to use instead of server-client mode.
+  This accelerate the get and put when no KVCache is matched. This version includes breaking API changes and is not backward compatible. 
+- add evict_ratio, need set "evict_ratio": 0.05 in cache_config
+- reducing the bubble inner the launch kernel
+- add vLLM 0.10.1.1 adapter
+
+### Fixed
+- cython release package
+
+
+## [0.1.0] - 2025-08-29
+
+### Init
+- init version
+- add license
+

VERSION

@@ -0,0 +1 @@
+1.0.0

docs/vllm_adapter/README_en.md

@@ -6,15 +6,15 @@ In commit [`0290841dce65ae9b036a23d733cf94e47e814934`](https://github.com/taco-p
 
 This change involves significant API adjustments. Therefore, please note:
 
-- **Version >= `0.0.2`**: Use the **current version API**; the vLLM patch is located in `examples/vllm_adaption/`.
-- **Version == `0.0.1`**: Supports the **legacy version API**; the vLLM patch is located in `examples/vllm_adaption_legacy/`.
+- **Version >= `1.0.0`**: Use the **current version API**; the vLLM patch is located in `examples/vllm_adaption/`.
+- **Version == `0.1.0`**: Supports the **legacy version API**; the vLLM patch is located in `examples/vllm_adaption_legacy/`.
 
 ---
 
-## Current Version (>= 0.0.2)
+## Current Version (>= 1.0.0)
 
 ### Supported Versions
-- FlexKV >= `0.0.2`
+- FlexKV >= `1.0.0`
 - vLLM versions >= `0.8.5` can generally follow this version for adaptation
 
 ### Example
@@ -63,10 +63,10 @@ VLLM_USE_V1=1 python -m vllm.entrypoints.cli.main serve Qwen3/Qwen3-32B \
 
 ```
 
-## Legacy Version (<= 0.0.1) – Not Recommended for Current Use
+## Legacy Version (<= 0.1.0) – Not Recommended for Current Use
 
 ### Supported Versions
-- FlexKV <= `0.0.1`
+- FlexKV <= `0.1.0`
 
 ### Example
 Apply the patch `examples/vllm_adaption_legacy/flexkv_vllm_0_8_4.patch` to vLLM 0.8.4, then start FlexKV, vLLM, and the benchmark script:

docs/vllm_adapter/README_zh.md

@@ -5,15 +5,15 @@
  **FlexKV 从 client-server 模式，变为推理加速引擎（如 vLLM）可直接调用的库函数**，以减少进程间消息传递的开销。
 这一变更引发了较大的 API 调整。因此，请注意：
 
-- **版本 >= `0.0.2`**：应使用 **当前版本 API**，vLLM patch位于 `examples/vllm_adaption/`。
-- **版本 == `0.0.1`**：仅支持 **Legacy 版本 API**, vLLM patch位于`examples/vllm_adaption_legacy/`。
+- **版本 >= `1.0.0`**：应使用 **当前版本 API**，vLLM patch位于 `examples/vllm_adaption/`。
+- **版本 == `0.1.0`**：仅支持 **Legacy 版本 API**, vLLM patch位于`examples/vllm_adaption_legacy/`。
 
 ---
 
-## 当前版本（>= 0.0.2）
+## 当前版本（>= 1.0.0）
 
 ### 适用版本
-- FlexKV >= `0.0.2`
+- FlexKV >= `1.0.0`
 - vLLM 原则上>= `0.8.5`版本均可参考示例代码进行修改
 
 ### 示例
@@ -62,10 +62,10 @@ VLLM_USE_V1=1 python -m vllm.entrypoints.cli.main serve Qwen3/Qwen3-32B \
 
 ```
 
-## Legacy版本（<= 0.0.1）,目前的版本尽量不要使用
+## Legacy版本（<= 0.1.0）,目前的版本尽量不要使用
 
 ### 适用版本
-- FlexKV <= `0.0.1`
+- FlexKV <= `0.1.0`
 
 ### 示例
 在 vLLM 0.8.4 版本中应用patch `examples/vllm_adaption_legacy/flexkv_vllm_0_8_4.patch`，分别启动 FlexKV、vLLM 和测试脚本：

setup.py

@@ -7,6 +7,9 @@
 from setuptools.command.build_ext import build_ext
 from torch.utils import cpp_extension
 
+def get_version():
+    with open(os.path.join(os.path.dirname(__file__), "VERSION")) as f:
+        return f.read().strip()
 
 build_dir = "build"
 os.makedirs(build_dir, exist_ok=True)
@@ -130,7 +133,7 @@ def copy_shared_libraries(self):
 setup(
     name="flexkv",
     description="A global KV-Cache manager for LLM inference",
-    version="0.1.0",
+    version=get_version(),
     packages=find_packages(exclude=("benchmarks", "csrc", "examples", "tests")),
     package_data={
         "flexkv": ["*.so", "lib/*.so", "lib/*.so.*"],