[PR #57] [MERGED] [feature] release 0.1.3 #1336

New issue

Closed

opened 2026-10-29 14:51:31 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-29 14:51:31 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/kvcache-ai/ktransformers/pull/57
Author: @UnicornChan
Created: 8/28/2024
Status: ✅ Merged
Merged: 8/28/2024
Merged by: @UnicornChan

Base: main ← Head: develop-0.1.3

📝 Commits (1)

4d1d561 [feature] release 0.1.3

📊 Changes

58 files changed (+11705 additions, -370 deletions)

View changed files

📝 .github/workflows/package_wheel_release.yml (+5 -29)
📝 .gitignore (+2 -1)
📝 README.md (+73 -25)
➕ doc/assets/Framework_effect.png (+0 -0)
➕ doc/assets/InfLLM_equation.jpg (+0 -0)
➕ doc/assets/InfLLM_framework.png (+0 -0)
➕ doc/assets/KTransformers_long_context_v1.png (+0 -0)
➕ doc/assets/KTransformers_long_context_v2.png (+0 -0)
➕ doc/assets/Quest_framework.png (+0 -0)
➕ doc/assets/SnapKV_framework.png (+0 -0)
➕ doc/assets/SparQ_attention.png (+0 -0)
➕ doc/assets/internlm_memory.png (+0 -0)
➕ doc/assets/long_context_generate.png (+0 -0)
➕ doc/assets/long_context_prefill.png (+0 -0)
➕ doc/assets/needle_128K.png (+0 -0)
➕ doc/assets/needle_1M.png (+0 -0)
➕ doc/en/long_context_tutorial.md (+316 -0)
📝 ktransformers/__init__.py (+11 -1)
📝 ktransformers/configs/config.yaml (+17 -1)
📝 ktransformers/ktransformers_ext/CMakeLists.txt (+7 -2)

...and 38 more files

📄 Description

support internlm2.5 for 1M Prompt under 24GB VRAM and 150GB DRAM
decrease DeepseekV2's required DRAM from 20G to 10G.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/kvcache-ai/ktransformers/pull/57 **Author:** [@UnicornChan](https://github.com/UnicornChan) **Created:** 8/28/2024 **Status:** ✅ Merged **Merged:** 8/28/2024 **Merged by:** [@UnicornChan](https://github.com/UnicornChan) **Base:** `main` ← **Head:** `develop-0.1.3` --- ### 📝 Commits (1) - [`4d1d561`](https://github.com/kvcache-ai/ktransformers/commit/4d1d561d28e04beb56f2c75b4ddaaf20d787ba07) [feature] release 0.1.3 ### 📊 Changes **58 files changed** (+11705 additions, -370 deletions) <details> <summary>View changed files</summary> 📝 `.github/workflows/package_wheel_release.yml` (+5 -29) 📝 `.gitignore` (+2 -1) 📝 `README.md` (+73 -25) ➕ `doc/assets/Framework_effect.png` (+0 -0) ➕ `doc/assets/InfLLM_equation.jpg` (+0 -0) ➕ `doc/assets/InfLLM_framework.png` (+0 -0) ➕ `doc/assets/KTransformers_long_context_v1.png` (+0 -0) ➕ `doc/assets/KTransformers_long_context_v2.png` (+0 -0) ➕ `doc/assets/Quest_framework.png` (+0 -0) ➕ `doc/assets/SnapKV_framework.png` (+0 -0) ➕ `doc/assets/SparQ_attention.png` (+0 -0) ➕ `doc/assets/internlm_memory.png` (+0 -0) ➕ `doc/assets/long_context_generate.png` (+0 -0) ➕ `doc/assets/long_context_prefill.png` (+0 -0) ➕ `doc/assets/needle_128K.png` (+0 -0) ➕ `doc/assets/needle_1M.png` (+0 -0) ➕ `doc/en/long_context_tutorial.md` (+316 -0) 📝 `ktransformers/__init__.py` (+11 -1) 📝 `ktransformers/configs/config.yaml` (+17 -1) 📝 `ktransformers/ktransformers_ext/CMakeLists.txt` (+7 -2) _...and 38 more files_ </details> ### 📄 Description 1. support internlm2.5 for 1M Prompt under 24GB VRAM and 150GB DRAM 2. decrease DeepseekV2's required DRAM from 20G to 10G. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-29 14:51:31 +00:00