Skip to content

Commit 0a9d72b

Browse files
committed
update docs
1 parent b1a667d commit 0a9d72b

File tree

6 files changed

+12
-6
lines changed

6 files changed

+12
-6
lines changed

README.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -33,6 +33,8 @@ docker pull registry.cn-hangzhou.aliyuncs.com/yongyang/llmcompression:pure-lates
3333

3434
## :fire: Latest News
3535

36+
- **August 13, 2025:** 🚀 We have open-sourced our compression solution for **vision-language models (VLMs)**, supporting over a total of **20 algorithms** that cover both **token reduction** and **quantization**. This release enables flexible, plug-and-play compression strategies for a wide range of multimodal tasks. please refer to the [documentation](https://llmc-en.readthedocs.io/en/latest/advanced/token_reduction.html).
37+
3638
- **May 12, 2025:** 🔥 We now fully support quantization for the **`Wan2.1`** series of video generation models and provide export of truly quantized **INT8/FP8** weights, compatible with the [lightx2v](https://github.com/ModelTC/lightx2v) inference framework. For details, please refer to the [lightx2v documentation](https://llmc-en.readthedocs.io/en/latest/backend/lightx2v.html).
3739

3840
- **Feb 07, 2025:** 🔥 We now fully support quantization of large-scale **`MOE`** models like **`DeepSeekv3`**, **`DeepSeek-R1`**, and **`DeepSeek-R1-zero`** with **`671B`** parameters. You can now directly load FP8 weights without any extra conversion. AWQ and RTN quantization can run on a single 80GB GPU, and we also support the export of true quantized **INT4/INT8** weights.

README_zh.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -33,6 +33,8 @@ docker pull registry.cn-hangzhou.aliyuncs.com/yongyang/llmcompression:pure-lates
3333

3434
## :fire: 最新动态
3535

36+
- **2025年8月13日:** 🚀 我们已开源针对 **视觉语言模型(VLMs)** 的压缩方案,支持共计超过 **20 种算法**,涵盖 **token reduction****quantization**。此次发布为多模态任务提供了灵活、即插即用的压缩策略。具体请参阅[文档](https://llmc-zhcn.readthedocs.io/en/latest/advanced/token_reduction.html)
37+
3638
- **2025年5月12日:** 🔥 我们现已全面支持 **`Wan2.1`** 系列视频生成模型的量化,并支持导出真实量化的 **INT8/FP8** 权重,兼容 [lightx2v](https://github.com/ModelTC/lightx2v) 推理框架。详情请参考 [lightx2v 使用文档](https://llmc-zhcn.readthedocs.io/en/latest/backend/lightx2v.html)
3739

3840
- **2025年2月7日:** 🔥 我们现已全面支持 **`DeepSeekv3`****`DeepSeek-R1`****`DeepSeek-R1-zero`** 等 671B 大规模 **`MOE`** 模型的量化。 您可以直接加载 `FP8` 权重,无需额外转换,使用单张 80G 显存的 GPU 即可运行 `AWQ``RTN` 量化,同时还支持导出真实量化的 **INT4/INT8** 权重

docs/en/source/index.rst

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -36,6 +36,7 @@ arxiv: https://arxiv.org/abs/2405.06001
3636
advanced/VLM_quant&img-txt_dataset.md
3737
advanced/mix_bits.md
3838
advanced/sparsification.md
39+
advanced/token_reduction.md
3940

4041
.. toctree::
4142
:maxdepth: 2

docs/zh_cn/source/advanced/VLM_quant&img-txt_dataset.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# VLM quant and custom_mm datatsets
1+
# VLM 量化和 custom_mm 数据集
22

33
llmc目前支持对VLM模型使用图像-文本数据集进行校准并量化
44

5-
## VLM quant
5+
## VLM 量化
66
当前支持的模型如下:
77
1. llava
88

@@ -34,7 +34,7 @@ calib:
3434
padding: True
3535
```
3636
37-
## custom_mm datatsets
37+
## custom_mm 数据集
3838
custom_mm 数据集格式如下:
3939
```
4040
custom_mm-datasets/

docs/zh_cn/source/advanced/Vit_quant&img_dataset.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# Vit quant and img datatsets
1+
# Vit 量化和 img 数据集
22

33
llmc目前支持对Vit模型使用图像数据集进行校准并量化
44

5-
## Vit quant
5+
## Vit 量化
66

77
下面是一个配置的例子
88

@@ -33,7 +33,7 @@ eval:
3333
eval_token_consist: False
3434
```
3535
36-
## img datatsets
36+
## img 数据集
3737
img数据集格式要求:img数据集目录下存在图像
3838
3939
img数据集格式示例:

docs/zh_cn/source/index.rst

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -37,6 +37,7 @@ arxiv链接: https://arxiv.org/abs/2405.06001
3737
advanced/VLM_quant&img-txt_dataset.md
3838
advanced/mix_bits.md
3939
advanced/sparsification.md
40+
advanced/token_reduction.md
4041

4142
.. toctree::
4243
:maxdepth: 2

0 commit comments

Comments
 (0)