Merge pull request #57 from UnicornChan/develop-0.1.3

[feature] release 0.1.3

Merge pull request #57 from UnicornChan/develop-0.1.3
[feature] release 0.1.3
233bbb8c · UnicornChan · GitHub · 67f8b370 · 4d1d561d · 233bbb8c
Unverified Commit 233bbb8c authored Aug 29, 2024 by UnicornChan Committed by GitHub Aug 29, 2024
20 changed files
--- a/.github/workflows/package_wheel_release.yml
+++ b/.github/workflows/package_wheel_release.yml
@@ -29,11 +29,6 @@ jobs:
         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.12', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
@@ -52,12 +47,6 @@ jobs:
         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
-         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.11', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
@@ -76,12 +65,6 @@ jobs:
         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
-         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'FANCY', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: ubuntu-20.04, pyver:  '3.10', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
@@ -98,10 +81,6 @@ jobs:
         - { os: windows-2022, pyver:  '3.12', cuda: '12.2.2', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.12', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.12', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
-         - { os: windows-2022, pyver:  '3.12', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.12', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.12', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.12', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
         - { os: windows-2022, pyver:  '3.12', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.12', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.12', cuda: '12.1.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
@@ -114,10 +93,6 @@ jobs:
         - { os: windows-2022, pyver:  '3.11', cuda: '12.2.2', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.11', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.11', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
-         - { os: windows-2022, pyver:  '3.11', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.11', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.11', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.11', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
         - { os: windows-2022, pyver:  '3.11', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.11', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.11', cuda: '12.1.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
@@ -130,10 +105,6 @@ jobs:
         - { os: windows-2022, pyver:  '3.10', cuda: '12.2.2', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.10', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.10', cuda: '12.1.1', torch: '2.4.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
-         - { os: windows-2022, pyver:  '3.10', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.10', cuda: '12.5.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.10', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '124'}
-         - { os: windows-2022, pyver:  '3.10', cuda: '12.4.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '124'}
         - { os: windows-2022, pyver:  '3.10', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.10', cuda: '12.2.2', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX2', torch_cu: '121'}
         - { os: windows-2022, pyver:  '3.10', cuda: '12.1.1', torch: '2.3.0', cudaarch: '8.0;8.6;8.7;8.9;9.0+PTX', instruct: 'AVX512', torch_cu: '121'}
@@ -219,6 +190,11 @@ jobs:
            $env:CUDA_PATH = "$env:CUDA_PATH/Library"
            $env:CUDA_HOME = $env:CUDA_PATH
            $env:PATH = "$env:CUDA_PATH/bin;" + $env:PATH
+            $directory = "$env:CUDA_PATH/lib/x64/"
+            if (-not (Test-Path -Path $directory)) {
+              New-Item -ItemType Directory -Path $directory
+              Write-Output "Directory '$directory' created."
+            }
            cp $env:CUDA_PATH/lib/*.lib $env:CUDA_PATH/lib/x64/
            $env:INCLUDE =$env:CUDA_PATH + "/include/targets/x64;" + $env:INCLUDE


--- a/.gitignore
+++ b/.gitignore
@@ -18,3 +18,4 @@ compile_commands.json
 ktransformers/server/local_store/
 ktransformers/server_test1.db
 *.patch
+local_chat_djw.py
\ No newline at end of file
--- a/README.md
+++ b/README.md
@@ -2,17 +2,16 @@
  <!-- <h1>KTransformers</h1> -->
  <p align="center">

-  <picture>
+<picture>
    <img alt="KTransformers" src="https://github.com/user-attachments/assets/d5a2492f-a415-4456-af99-4ab102f13f8b" width=50%>

-  </picture>
+</picture>

-  </p>
+</p>
  <h3>A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations</h3>
  <strong><a href="#show-cases">🔥 Show Cases</a> | <a href="#quick-start">🚀 Quick Start</a> | <a href="#tutorial">📃 Tutorial</a> | <a href="https://github.com/kvcache-ai/ktransformers/discussions">💬  Discussion </a> </strong>
 </div>

-
 <h2 id="intro">🎉 Introduction</h2>
 KTransformers, pronounced as Quick Transformers, is designed to enhance your 🤗 <a href="https://github.com/huggingface/transformers">Transformers</a> experience with advanced kernel optimizations and placement/parallelism strategies.
 <br/><br/>
@@ -22,16 +21,42 @@ interface, RESTful APIs compliant with OpenAI and Ollama, and even a simplified
 <br/><br/>
 Our vision for KTransformers is to serve as a flexible platform for experimenting with innovative LLM inference optimizations. Please let us know if you need any other features.

-<h2 id="Updates">✨ Updates</h2>
+<h2 id="Updates">🔥 Updates</h2>

+* **Aug 28, 2024**: Support 1M context under the InternLM2.5-7B-Chat-1M model, utilizing 24GB of VRAM and 150GB of DRAM.
+* **Aug 28, 2024**: Decrease DeepseekV2's required DRAM from 20G to 10G.
 * **Aug 15, 2024**: Update detailed [TUTORIAL](doc/en/injection_tutorial.md) for injection and multi-GPU. 
-* **Aug 14, 2024**: Support llamfile as linear backend, 
+* **Aug 14, 2024**: Support llamfile as linear backend. 
 * **Aug 12, 2024**: Support multiple GPU; Support new model: mixtral 8\*7B  and 8\*22B; Support q2k, q3k, q5k dequant on gpu.
 * **Aug 9, 2024**: Support windows native.

 <h2 id="show-cases">🔥 Show Cases</h2>
-<h3>GPT-4-level Local VSCode Copilot on a Desktop with only 24GB VRAM</h3>
+<h3>1M Context Local Inference on a Desktop with Only 24GB VRAM</h3>
+<p align="center">
+
+https://github.com/user-attachments/assets/a865e5e4-bca3-401e-94b8-af3c080e6c12
+
+* **1M Context InternLM 2.5 7B**: Operates at full bf16 precision, utilizing 24GB VRAM and 150GB DRAM, which is feasible on a local desktop setup. It achieves a 92.88% success rate on the 1M "Needle In a Haystack" test and 100% on the 128K NIAH test.
+
+<p align="center">
+  <picture>
+    <img alt="Single Needle Retrieval 128K" src="./doc/assets/needle_128K.png" width=100%>
+  </picture>
+</p>
+
 <p align="center">
+  <picture>
+    <img alt="Single Needle Retrieval 1000K" src="./doc/assets/needle_1M.png" width=100%>
+  </picture>
+</p>
+
+* **Enhanced Speed**: Reaches 16.91 tokens/s for generation with a 1M context using sparse attention, powered by llamafile kernels. This method is over 10 times faster than full attention approach of llama.cpp.
+
+* **Flexible Sparse Attention Framework**: Offers a flexible block sparse attention framework for CPU offloaded decoding. Compatible with SnapKV, Quest, and InfLLm. Further information is available [here](./doc/en/long_context_tutorial.md).
+
+<div>
+<h3>GPT-4-level Local VSCode Copilot on a Desktop with only 24GB VRAM</h3>
+</div>

 https://github.com/user-attachments/assets/0b9fa2da-66f0-48eb-b4b9-f0e1f06f8927

@@ -54,7 +79,6 @@ https://github.com/user-attachments/assets/4c6a8a38-05aa-497d-8eb1-3a5b3918429c

 </p>

-
 <strong>More advanced features will coming soon, so stay tuned!</strong>

 <h2 id="quick-start">🚀 Quick Start</h2>
@@ -89,17 +113,21 @@ Some preparation:
  ```
  
 - Linux-x86_64 with gcc, g++ and cmake
+  
  ```sh
  sudo apt-get update
  sudo apt-get install gcc g++ cmake ninja-build
  ```
+
 - We recommend using [Conda](https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh) to create a virtual environment with Python=3.11 to run our program.
+  
  ```sh
  conda create --name ktransformers python=3.11
  conda activate ktransformers # you may need to run ‘conda init’ and reopen shell first
  ```

 - Make sure that PyTorch, packaging, ninja is installed
+  
  ```
  pip install torch packaging ninja
  ```
@@ -107,28 +135,36 @@ Some preparation:
 <h3>Installation</h3>

 1. Use a Docker image, see [documentation for Docker](./doc/en/docker.md) 
+
 2. You can install using Pypi (for linux):
   
   ```
   pip install ktransformers --no-build-isolation
   ```
+   
   for windows we prepare a pre compiled whl package in [ktransformers-0.1.1+cu125torch24avx2-cp311-cp311-win_amd64.whl](https://github.com/kvcache-ai/ktransformers/releases/download/v0.1.1/ktransformers-0.1.1+cu125torch24avx2-cp311-cp311-win_amd64.whl), which require cuda-12.5, torch-2.4, python-3.11, more pre compiled package are being produced. 

 3. Or you can download source code and compile:
+   
   - init source code 
+     
     ```sh
     git clone https://github.com/kvcache-ai/ktransformers.git
     cd ktransformers
     git submodule init
     git submodule update
     ```
+   
   - [Optional] If you want to run with website, please [compile the website](./doc/en/api/server/website.md) before execute ```bash install.sh```
+   
   - Compile and install (for Linux)
+     
     ```
     bash install.sh
     ```
   
   - Compile and install(for Windows)
+     
     ```
     install.bat
     ```
@@ -136,8 +172,7 @@ Some preparation:
 <h3>Local Chat</h3>
 We provide a simple command-line local chat Python script that you can run for testing.

-  > Note that this is a very simple test tool only support one round chat without any memory about last input, if you want to try full ability of the model, you may go to [RESTful API and Web UI](#id_666). We use the DeepSeek-V2-Lite-Chat-GGUF model as an example here. But we also support other models, you can replace it with any other model that you want to test. 
-
+> Note that this is a very simple test tool only support one round chat without any memory about last input, if you want to try full ability of the model, you may go to [RESTful API and Web UI](#id_666). We use the DeepSeek-V2-Lite-Chat-GGUF model as an example here. But we also support other models, you can replace it with any other model that you want to test. 

 <h4>Run Example</h4>

@@ -162,23 +197,30 @@ python -m ktransformers.local_chat --model_path deepseek-ai/DeepSeek-V2-Lite-Cha
 # python  ktransformers.local_chat --model_path ./DeepSeek-V2-Lite --gguf_path ./DeepSeek-V2-Lite-Chat-GGUF
 ```

-
 It features the following arguments:

 - `--model_path` (required): Name of the model (such as "deepseek-ai/DeepSeek-V2-Lite-Chat" which will automatically download configs from [Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-V2-Lite)). Or if you already got local files  you may directly use that path to initialize the model.  
-  >Note: <strong>.safetensors</strong> files are not required in the directory. We only need config files to build model and tokenizer.
+  
+  > Note: <strong>.safetensors</strong> files are not required in the directory. We only need config files to build model and tokenizer.
+
 - `--gguf_path` (required): Path of a directory containing GGUF files which could that can be downloaded from [Hugging Face](https://huggingface.co/mzwing/DeepSeek-V2-Lite-Chat-GGUF/tree/main) (we only support q4_k_m and q8_0 for now, more formats are coming soon).
+
 - `--optimize_rule_path` (required except for Qwen2Moe and DeepSeek-V2): Path of YAML file containing optimize rules. There are two rule files pre-written in the [ktransformers/optimize/optimize_rules](ktransformers/optimize/optimize_rules) directory for optimizing DeepSeek-V2 and Qwen2-57B-A14, two SOTA MoE models.
+
 - `--max_new_tokens`: Int (default=1000). Maximum number of new tokens to generate.
+
 - `--cpu_infer`: Int (default=10). The number of CPUs used for inference. Should ideally be set to the (total number of cores - 2).

 <h3 id="supported-model"> Supported Model</h3>

 | Model Name                     | Model Size | VRAM  | Minimum DRAM    | Recommended DRAM  |
-| ----  | ---- | ---- | ---- | ---- |
-| DeepSeek-V2-q4_k_m | 133G | 24G | 136G | 192G |
+| ------------------------------ | ---------- | ----- | --------------- | ----------------- |
+| DeepSeek-V2-q4_k_m             | 133G       | 10G   | 136G            | 192G              |
 | Qwen2-57B-A14B-Instruct-q4_k_m | 33G        | 8G    | 34G             | 64G               |
 | DeepSeek-V2-Lite-q4_k_m        | 9.7G       | 3G    | 13G             | 16G               |
+| Mixtral-8x7B-q4_k_m            | 25G        | 1.6G  | 51G             | 64G               |
+| Mixtral-8x22B-q4_k_m           | 80G        | 4G    | 86.1G           | 96G               |
+| InternLM2.5-7B-Chat-1M         | 15.5G      | 15.5G | 8G(32K context) | 150G (1M context) |


 More will come soon. Please let us know which models you are most interested in. 
@@ -188,7 +230,6 @@ Be aware that you need to be subject to their corresponding model licenses when
 <details>
  <summary>Click To Show how to run other examples</summary>

-
 * Qwen2-57B

 ```sh
@@ -208,6 +249,7 @@ python -m ktransformers.local_chat --model_name Qwen/Qwen2-57B-A14B-Instruct --g
 ```

 * DeepseekV2
+  
 ```sh
 mkdir DeepSeek-V2-Chat-0628-GGUF && cd DeepSeek-V2-Chat-0628-GGUF
 # Download weights
@@ -221,8 +263,11 @@ cd ..
 python -m ktransformers.local_chat --model_name deepseek-ai/DeepSeek-V2-Chat-0628 --gguf_path ./DeepSeek-V2-Chat-0628-GGUF

 # If you see “OSError: We couldn't connect to 'https://huggingface.co' to load this file”, try：
+
 # GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/deepseek-ai/DeepSeek-V2-Chat-0628
+
 # python -m ktransformers.local_chat --model_path ./DeepSeek-V2-Chat-0628 --gguf_path ./DeepSeek-V2-Chat-0628-GGUF
+
 ```

 | model name | weights download link |
@@ -245,11 +290,15 @@ Start without website:
 ```sh
 ktransformers --model_path deepseek-ai/DeepSeek-V2-Lite-Chat --gguf_path /path/to/DeepSeek-V2-Lite-Chat-GGUF --port 10002
 ```
+
 Start with website:
+
 ```sh
 ktransformers --model_path deepseek-ai/DeepSeek-V2-Lite-Chat --gguf_path /path/to/DeepSeek-V2-Lite-Chat-GGUF  --port 10002 --web True
 ```
+
 Or you want to start server with transformers, the model_path should include safetensors
+
 ```bash
 ktransformers --type transformers --model_path /mnt/data/model/Qwen2-0.5B-Instruct --port 10002 --web True
 ```
@@ -264,7 +313,6 @@ Access website with url [http://localhost:10002/web/index.html#/chat](http://loc

 More information about the RESTful API server can be found [here](doc/en/api/server/server.md). You can also find an example of integrating with Tabby [here](doc/en/api/server/tabby.md).

-
 <h2 id="tutorial">📃 Brief Injection Tutorial</h2>
 At the heart of KTransformers is a user-friendly, template-based injection framework. 
 This allows researchers to easily replace original torch modules with optimized variants. It also simplifies the process of combining multiple optimizations, allowing the exploration of their synergistic effects.

--- a/doc/assets/Framework_effect.png
+++ b/doc/assets/Framework_effect.png
--- a/doc/assets/InfLLM_equation.jpg
+++ b/doc/assets/InfLLM_equation.jpg
--- a/doc/assets/InfLLM_framework.png
+++ b/doc/assets/InfLLM_framework.png
--- a/doc/assets/KTransformers_long_context_v1.png
+++ b/doc/assets/KTransformers_long_context_v1.png
--- a/doc/assets/KTransformers_long_context_v2.png
+++ b/doc/assets/KTransformers_long_context_v2.png
--- a/doc/assets/Quest_framework.png
+++ b/doc/assets/Quest_framework.png
--- a/doc/assets/SnapKV_framework.png
+++ b/doc/assets/SnapKV_framework.png
--- a/doc/assets/SparQ_attention.png
+++ b/doc/assets/SparQ_attention.png
--- a/doc/assets/internlm_memory.png
+++ b/doc/assets/internlm_memory.png
--- a/doc/assets/long_context_generate.png
+++ b/doc/assets/long_context_generate.png
--- a/doc/assets/long_context_prefill.png
+++ b/doc/assets/long_context_prefill.png
--- a/doc/assets/needle_128K.png
+++ b/doc/assets/needle_128K.png
--- a/doc/assets/needle_1M.png
+++ b/doc/assets/needle_1M.png
--- a/doc/en/long_context_tutorial.md
+++ b/doc/en/long_context_tutorial.md
--- a/ktransformers/__init__.py
+++ b/ktransformers/__init__.py
-__version__ = "0.1.2"
\ No newline at end of file
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Description  : 
+Author       : kkk1nak0
+Date         : 2024-08-15 07:34:46
+Version      : 1.0.0
+LastEditors  : chenxl 
+LastEditTime : 2024-08-28 15:19:03
+'''
+__version__ = "0.1.3"
--- a/ktransformers/configs/config.yaml
+++ b/ktransformers/configs/config.yaml
@@ -35,3 +35,19 @@ web:

 ext:
  cpu_infer: 10
+
+long_context:
+  chunk_size: 4096
+  max_seq_len: 32000
+  block_size: 128
+  local_windows_len: 4096
+  second_select_num: 32
+  anchor_type: DYNAMIC
+  kv_type: FP16
+  dense_layer_num: 2
+  anchor_num: 1
+  preselect_block: True
+  head_select_mode: SHARED
+  preselect_block_count: 32
+  layer_step: 1
+  token_step: 100
\ No newline at end of file
--- a/ktransformers/ktransformers_ext/CMakeLists.txt
+++ b/ktransformers/ktransformers_ext/CMakeLists.txt
 cmake_minimum_required(VERSION 3.16)
 project(cpuinfer_ext VERSION 0.1.0)

+
 set(CMAKE_CXX_STANDARD 17)
 set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3 -ffast-math")
 set(CMAKE_BUILD_TYPE "Release")
@@ -215,7 +216,8 @@ aux_source_directory(${CMAKE_CURRENT_SOURCE_DIR} SOURCE_DIR1)
 aux_source_directory(${CMAKE_CURRENT_SOURCE_DIR}/cpu_backend SOURCE_DIR2)
 aux_source_directory(${CMAKE_CURRENT_SOURCE_DIR}/operators/llamafile SOURCE_DIR3)
 aux_source_directory(${CMAKE_CURRENT_SOURCE_DIR}/../../third_party/llamafile SOURCE_DIR4)
-set(ALL_SOURCES ${SOURCE_DIR1} ${SOURCE_DIR2} ${SOURCE_DIR3} ${SOURCE_DIR4})
+aux_source_directory(${CMAKE_CURRENT_SOURCE_DIR}/operators/kvcache SOURCE_DIR5)
+set(ALL_SOURCES ${SOURCE_DIR1} ${SOURCE_DIR2} ${SOURCE_DIR3} ${SOURCE_DIR4} ${SOURCE_DIR5})
 message(STATUS "ALL_SOURCES: ${ALL_SOURCES}")

 pybind11_add_module(${PROJECT_NAME} MODULE ${ALL_SOURCES})
@@ -223,5 +225,8 @@ target_link_libraries(${PROJECT_NAME} PRIVATE llama)
 if(WIN32)
    target_link_libraries(${PROJECT_NAME} PRIVATE "$ENV{CUDA_PATH}/lib/x64/cudart.lib")#CUDA::cudart
 elseif(UNIX)
+    if(NOT DEFINED ENV{CUDA_HOME} OR "$ENV{CUDA_HOME}" STREQUAL "")
+        set(ENV{CUDA_HOME} "/usr/local/cuda")
+    endif()
    target_link_libraries(${PROJECT_NAME} PRIVATE "$ENV{CUDA_HOME}/lib64/libcudart.so")
 endif()