Skip to content

GitLab

Menu

Projects Groups Snippets

Loading...
Help
Sign in / Register

M ModelZoo
Group information
Issues 41
Merge requests 6
- Merge requests 6
Packages & Registries

Collapse sidebar

Collapse sidebar

ModelZoo
Issues

Open 41
Closed 135
All 176

Priority Created date Last updated Milestone due date Due date Popularity Label priority Manual

编译报错

llama_fastertransformer#7 · created May 14, 2024 by xurui
- CLOSED
- 2
updated Nov 05, 2024
K100AI用24.04.1镜像运行qwen2-72b无法启动

qwen1.5_vllm#3 · created Jul 02, 2024 by jixx
- CLOSED
- 1
updated Nov 05, 2024
vllm能否做个webserver的推理例子

llama_vllm#1 · created Jun 21, 2024 by ncic_liuyao
- CLOSED
- 3
updated Oct 21, 2024
Qwen2-72B-Instruct-GPTQ-Int4测试benchmark_throughput,在固定输入输出长度下，随着num-prompts增加没有线性能力

qwen1.5_vllm#4 · created Aug 01, 2024 by xurui
- CLOSED
- 2
updated Oct 16, 2024
K100上报错CUDA error: HIPBLAS_STATUS_NOT_SUPPORTED

qwen1.5_vllm#2 · created Jun 14, 2024 by xurui
- CLOSED
- 1
updated Oct 16, 2024
glm4 9b 推理都是感叹号！！！

chatglm_vllm#1 · created Jul 10, 2024 by quyuanhao123
- CLOSED
- 0
updated Oct 16, 2024
使用 CodeLlama-7b-Instruct-hf 推理，生成文本为噪音

llama_vllm#5 · created Jul 04, 2024 by acane
- CLOSED
- 2
updated Oct 16, 2024
使用llama-7b-chat推理测试，得到答案不合理

llama_vllm#4 · created Jul 02, 2024 by eating1
- CLOSED
- 2
updated Oct 16, 2024
vllm能否适配最新版本，为了能支持推理llama3

llama_vllm#2 · created Jun 21, 2024 by ncic_liuyao
- CLOSED
- 2
updated Oct 16, 2024
qwen2.5 7B跑8卡报错

qwen2.5_pytorch#1 · created Oct 15, 2024 by gaoxuan
- CLOSED
- 1
updated Oct 15, 2024
RuntimeError: [enforce fail at /data/jenkins_workspace/workspace/pytorch@4/third_party/gloo/gloo/transport/tcp/device.cc:210] ifa != nullptr. Unable to find interface for: [0.5.118.124]

internlm_2.5_pytorch#2 · created Sep 25, 2024 by xurui
- CLOSED
- 1
updated Oct 11, 2024
该项目没有包含Internlm_2.5工程的代码？

internlm_2.5_pytorch#1 · created Sep 24, 2024 by xurui
- CLOSED
- 0
updated Sep 25, 2024
readme中数据集下载链接失效

bert-pytorch#1 · created Sep 13, 2024 by xurui
- CLOSED
- 1
updated Sep 23, 2024
Z100下报错unsupported conversion from f16 to f16之后报错Aborted (core dumped)

stablediffusion_v2.1_pytorch#1 · created Aug 17, 2024 by zhangxp1
- CLOSED
- 4
updated Sep 13, 2024
报错ImportError: cannot import name 'is_torch_mlu_available' from 'transformers.utils' (/usr/local/lib/python3.10/site-packages/transformers/utils/__init__.py)

llama2_lora_pytorch#1 · created Sep 12, 2024 by xurui
- CLOSED
- 2
updated Sep 12, 2024
希望benchmark增加对dtk23.10的K100卡的支持

chatglm2-6b_fastllm#3 · created Mar 26, 2024 by youbo
- CLOSED
- 1
updated Aug 06, 2024
这个案例的yolov8采用的预处理好像并非是官方的letterbox？

yolov8_migraphx#2 · created Apr 26, 2024 by wangkaixiong
- CLOSED
- 1
updated Aug 06, 2024
当我编译模型的option设置offload_copy为false时，在migraphx上，推理结果始终会拷贝会cpu一份；

yolov8_migraphx#1 · created Apr 24, 2024 by wangkaixiong
- CLOSED
- 2
updated Aug 06, 2024
下载的模型为.safetensors文件，没有ckpt文件。

gemma_pytorch#1 · created Mar 21, 2024 by xurui
- CLOSED
- 1
updated Aug 06, 2024
希望能够补充一下inference或evaluation的步骤

qwen-torch#1 · created Oct 07, 2023 by Sugon_ldc
- CLOSED
- 3
updated Aug 06, 2024

Prev
1
2
3
4
5
6
7
Next