Commits · 3d328d61190cfbe44e992992ee520ccc2f80c88f · jerrrrry / infinilm

06 Dec, 2025 1 commit
- issue/92 添加InferEngine，支持多线程推理 · 3d328d61
  PanZezhong1725 authored Dec 06, 2025
  
  3d328d61
05 Dec, 2025 1 commit
- fix xmake · 0794f307
  PanZezhong authored Dec 05, 2025
  
  0794f307
04 Dec, 2025 8 commits
- Merge pull request #104 from InfiniTensor/issue/103 · f88a06f3
  PanZezhong1725 authored Dec 04, 2025
```
issue/103 新版分布式推理基建
```
  f88a06f3
- Merge pull request #101 from pengcheng888/issue/89 · 9dd10678
  pengcheng888 authored Dec 04, 2025
```
issue/89 在python的llama中使用matmul函数、以及减少Tensor对象创建次数
```
  9dd10678
- issue/103 新版分布式推理基建 · 1f5ab1c5
  PanZezhong authored Dec 04, 2025
  
  1f5ab1c5
- issue/89 针对python的llama结构，使用matmul函数、减少Tensor对象创建次数 · 7e59976b
  pengcheng888 authored Dec 04, 2025
  
  7e59976b
- Merge pull request #98 from InfiniTensor/issue/97 · 36f8eab7
  PanZezhong1725 authored Dec 04, 2025
```
issue/97 Attention 和 KVCache 支持 batch 维度
```
  36f8eab7
- Merge pull request #100 from InfiniTensor/issue/99 · 42f9d47d
  PanZezhong1725 authored Dec 04, 2025
```
issue/99 - relocated pybind contents
```
  42f9d47d
- issue/99 - relocated pybind contents · 3e398429
  wooway777 authored Dec 04, 2025
  
  3e398429
- issue/97 推理脚本支持batch · 6f624c94
  PanZezhong authored Dec 04, 2025
  
  6f624c94
03 Dec, 2025 4 commits
- issue/97 Attention 和 KVCache 支持 batch 维度 · 0da7b5db
  PanZezhong authored Dec 03, 2025
  
  0da7b5db
- Merge pull request #96 from InfiniTensor/issue/95 · a1f6e517
  PanZezhong1725 authored Dec 03, 2025
```
issue/95 将pybind target命名为_infinilm
```
  a1f6e517
- issue/95 将pybind target命名为_infinilm · 38ac084d
  PanZezhong authored Dec 03, 2025
  
  38ac084d
- Merge pull request #75 from InfiniTensor/issue/74 · ae3ebe19
  PanZezhong1725 authored Dec 03, 2025
```
Issue/74 基于InfiniCore::nn::module适配Llama模型
```
  ae3ebe19
02 Dec, 2025 1 commit
- issue/74 add c++ Llama models and align to AutoLlama interface · d6a641d3
  Ceng23333 authored Dec 02, 2025
```
Signed-off-by: Ceng23333 <441651826@qq.com>
```
  d6a641d3
26 Nov, 2025 5 commits
- Merge pull request #87 from InfiniTensor/issue/86 · 3c6ad521
  PanZezhong1725 authored Nov 26, 2025
```
issue/86 - 添加计算结束前的同步函数
```
  3c6ad521
- issue/86 - 添加计算结束前的同步函数 · 0a5f4d71
  pengcheng888 authored Nov 26, 2025
  
  0a5f4d71
- Merge pull request #68 from InfiniTensor/fix_std_move · efc19d7c
  PanZezhong1725 authored Nov 26, 2025
```
removed std::move
```
  efc19d7c
- Merge pull request #84 from pengcheng888/issue/83 · fa7ca75d
  PanZezhong1725 authored Nov 26, 2025
```
issue/83 - 添加AutoLlama类，支持创建不同backend的模型
```
  fa7ca75d
- issue/83 - 添加AutoLlama类，支持创建不同backend的模型 · e4114c03
  pengcheng888 authored Nov 26, 2025
  
  e4114c03
22 Nov, 2025 7 commits
- Merge pull request #82 from InfiniTensor/issue/80 · 5d182420
  PanZezhong1725 authored Nov 22, 2025
```
修复attention prefill计时方式，重构目录
```
  5d182420
- issue/80 增加对musa同步支持 · 02676be8
  PanZezhong authored Nov 22, 2025
  
  02676be8
- issue/80 修复attention prefill计时方式，重构目录 · d7d0889d
  PanZezhong authored Nov 22, 2025
  
  d7d0889d
- Merge pull request #81 from pengcheng888/issue/80 · 469f2884
  pengcheng888 authored Nov 22, 2025
```
issue/80 模型文件夹名字改为为model_path,增加llama模型的moore, iluvatar平台的参数
```
  469f2884
- issue/80 模型文件夹名字改为为model_path,增加moore平台的参数 · faaa9ec3
  pengcheng888 authored Nov 22, 2025
  
  faaa9ec3
- Merge pull request #79 from pengcheng888/issue/78 · 7d222d83
  pengcheng888 authored Nov 22, 2025
```
issue/78 - 在test文件夹目录,添加qwen3的moe和attention的耗时测试脚本
```
  7d222d83
- issue/78 - 在test文件夹目录,添加qwen3的moe和attention的耗时测试脚本 · d80bd14a
  pengcheng888 authored Nov 22, 2025
  
  d80bd14a
21 Nov, 2025 2 commits
- Merge pull request #77 from pengcheng888/main-python · 9e727b14
  pengcheng888 authored Nov 21, 2025
```
issue/76 - 添加python的llama模型实现
```
  9e727b14
- 调整repeat_kv的代码 · c53104e2
  pengcheng888 authored Nov 21, 2025
  
  c53104e2
20 Nov, 2025 1 commit
- issue/76 - 添加python的llama模型实现 · 3ddffe8d
  pengcheng888 authored Nov 20, 2025
  
  3ddffe8d
06 Nov, 2025 1 commit
- issue/71: support vllm benchmark text completion in InfiniLM · 4fd9d490
  spike-zhu authored Nov 06, 2025
  
  4fd9d490
03 Nov, 2025 1 commit
- removed std::move · 6fd5bb93
  wooway777 authored Nov 03, 2025
  
  6fd5bb93
29 Oct, 2025 1 commit
- Merge pull request #65 from InfiniTensor/issue/64 · 1c710c19
  PanZezhong1725 authored Oct 29, 2025
```
issue/64 - jiuge.py verbose output
```
  1c710c19
28 Oct, 2025 1 commit
- issue/64 - jiuge.py verbose output · 7a412203
  wooway777 authored Oct 28, 2025
  
  7a412203
22 Oct, 2025 1 commit
- Issue/60 -main 修复输出token乱码并适配了qwen3模型 · 80909bee
  pengcheng888 authored Oct 22, 2025
  
  80909bee
10 Oct, 2025 2 commits
- 增加issue模板 · 753a4f60
  PanZezhong authored Oct 10, 2025
  
  753a4f60
- feat: add ceval test script · 515b9245
  PanZezhong authored Oct 10, 2025
  
  515b9245
29 Sep, 2025 2 commits
- Merge pull request #46 from gongchensu/feature/support_hygon · ddea3d19
  PanZezhong1725 authored Sep 29, 2025
```
Compatible with Hygon DCU.
```
  ddea3d19
- Compatible with Hygon DCU. · 2a51ff9e
  zhuyue authored Sep 12, 2025
  
  2a51ff9e
24 Sep, 2025 1 commit
- Merge pull request #55 from InfiniTensor/issue/53 · d39a5e4d
  PanZezhong1725 authored Sep 24, 2025
```
issue/53: update AWQ-dequantize op name to match infinicore op dequantizeAWQ
```
  d39a5e4d