issues
search
QwenLM
/
Qwen
The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.
Apache License 2.0
13.59k
stars
1.11k
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add Docker image for CUDA-12.1.
#922
fyabc
closed
9 months ago
0
词表裁剪后编码单个词正常，重复单个词或多个词no entry found for key
#921
andrewhuman
closed
9 months ago
5
[BUG] RuntimeErrorExpected all tensors to be on the same device, but found at least two devices, cuda:2 and cuda:1!
#920
phoenixbai
closed
9 months ago
1
JSON format error in create_pandas_dataframe_agent
#919
Haxeebraja
closed
7 months ago
3
💡 [REQUEST] - 请问啥时候14b和7b可以修改system。最近用了72b修改system，非常不错。
#918
wellcasa
closed
9 months ago
2
Qwen14B-Chat-int4[BUG] <title>
#917
xx-Jiangwen
closed
9 months ago
0
[BUG] <title>单机8*32GV100 QLora微调 Qwen-72B-Chat-Int4仍然OOM
#916
15024287710Jackson
closed
6 months ago
5
How to eval the model performance during full-parameter finetuning?
#915
tiesanguaixia
closed
5 months ago
1
为什么14B-int4的推理时间比14B的慢很多
#914
boquanzhou
closed
9 months ago
8
vLLM还是不支持v100啊
#913
boquanzhou
closed
9 months ago
1
💡 [REQUEST] - 新增100M级别小模型（speculative decode加速使用）
#912
cermeng
closed
6 months ago
4
Qwen-7B-Chat-Int4微调后推理结果偏差有点大
#910
chesp
closed
5 months ago
3
qwen-14b精调是否需要外推
#909
LuoKaiGSW
closed
9 months ago
1
[BUG] <title> 1.8B 模型 system prompt 在部分情况下失效（ 如通过 fastchat.serve.openai_api_server ）
#908
1100111GTH
closed
9 months ago
3
构建新的docker镜像时报错 unknown instruction: APT
#907
haowuliao
closed
9 months ago
7
[BUG] <1.8b模型推理速度比7b慢>
#906
AIFFFENG
closed
9 months ago
3
使用FunctionCall时，生成Function的日期参数错误
#905
LiuGuBiGu
closed
5 months ago
2
[BUG] <title>Qwen-72B-Chat-Int4的本地模型加载打开失败
#904
15024287710Jackson
closed
9 months ago
1
请问一下能否开源一下工具调用的数据集呢，因为论文里写的感觉不是很清楚
#903
wangjvjie
closed
5 months ago
4
[BUG] <title> 使用示例代码提示缺少文件 pytorch_model.bin.index.json
#902
lianghongkey
closed
9 months ago
5
TypeError: QWenLMHeadModel.forward() got an unexpected keyword argument 'decoder_input_ids'[BUG] <title>
#901
AlexJJJChen
closed
9 months ago
1
[BUG] 四卡 3090 无法进行推理
#898
chopin1998
closed
9 months ago
1
[BUG] openai_api.py 使用 Qwen-7B-Chat-Int4，报错RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'
#897
LiuGuBiGu
closed
9 months ago
1
量化模型，如何分布式流式推理。
#896
YYGe01
closed
9 months ago
2
[BUG] 2张32GB显卡，对Qwen-14B进行lora微调，报错torch.cuda.0utOfMemoryError: CUDA outof memory
#895
Hazards10
closed
9 months ago
1
[BUG] <title>A800多卡执行微调脚本失败 使用镜像 qwenllm/qwen:cu117
#894
oho-work
closed
9 months ago
3
[BUG] Langchain Function Call Error
#893
Tejaswgupta
closed
9 months ago
3
[BUG] Qwen readme中介绍的批量推理方式失效
#892
tomFoxxxx
closed
9 months ago
2
Please support quip# 2bit quantization for the qwen model
#891
Minami-su
closed
9 months ago
5
💡 [REQUEST] - 咨询一下微调效果，使用gptq模型去微调和使用正常模型去微调后再转gptq，效果差异大不，假设微调数据量有1万多
#890
wengyuan722
closed
9 months ago
3
[BUG] Qwen-72B使用vLLM部署报告input prompt is too long
#889
TangMoming
closed
9 months ago
3
[BUG] <title>72B全量微调资源需求
#888
liyang-7
closed
9 months ago
1
[BUG] <title> Qwen-14B全量微调的显存使用量非常高
#887
hackiey
closed
5 months ago
7
[问题] <为什么lora微调base模型需要的显存比微调chat模型要高？>
#886
tangsipeng
closed
9 months ago
1
Qwen-72B-Chat-Int4 每次推理结果不一致
#885
lyc728
closed
9 months ago
1
采用cpu方式推理Qwen-1_8B-Chat-Int4 报错
#884
SiriusWy
closed
9 months ago
1
本地部署Qwen-7B之后如何开启API服务？💡 [REQUEST] - <求助>
#883
yoohhwz
closed
9 months ago
1
[BUG] <Qwen-72B-Chat-Int4 推理速度为什么比Qwen-72B-Chat慢很多>
#882
vipcong816
closed
9 months ago
3
Fix peft version in dockerfiles.
#881
fyabc
closed
9 months ago
0
使用int8量化qwen-14b模型后，首字响应时间相比int4量化的模型慢了很多
#880
Originhhh
closed
9 months ago
6
💡 [REQUEST] - <1.8B模型商用>
#879
yuimo
closed
9 months ago
1
💡 [REQUEST] - 能不能把微调语料的格式改为jsonl（Json Line）
#878
zhengyangyong
closed
9 months ago
2
[BUG] Qwen72B-Chat-Int4部署vLLM_gptq，API调用，任务推理过慢时会出现 Aborted request
#877
SL-Stone
closed
5 months ago
10
[BUG] Qwen72B-Chat-Int4部署vLLM_gptq，API调用时处理过慢，出现 Aborted request，以及Pending reqs过多的现象
#876
SL-Stone
closed
9 months ago
0
[BUG] docker run报错，
#875
chesp
closed
9 months ago
40
[BUG] <Qwen-72b-Chat lora+Deepspeed-3微调出现loss始终不下降的情况>
#874
long-night
closed
5 months ago
4
No module named 'langchain.tools.python.tool'
#873
huangj17
closed
9 months ago
3
[BUG] flash attention推理速度疑问
#872
Data2Me
closed
9 months ago
2
微调时报错We couldn't connect to 'https://huggingface.co'
#871
chesp
closed
9 months ago
1
[BUG] <title> vllm_wrapper如何设置多卡推理？
#870
ChenlongDeng
closed
9 months ago
2
Previous
Next