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
Qwen2.5 is here! See https://github.com/QwenLM/Qwen2.5
#1097
jklj077
closed
5 months ago
0
[BUG] 多轮对话数据微调训练,token_type_ids为None
#1096
sunyclj
closed
7 months ago
1
[BUG] <title> Gradio demo 无法正常显示模型输出
#1095
zzc20001
closed
5 months ago
3
请问新的1.5版本sft的lr为什么到1e-5后就不下降了呢
#1093
128Ghe980
closed
7 months ago
0
请问模型并发上限有没有测试
#1092
rabum
closed
6 months ago
2
[BUG] <title>When quantifying the trained Qwen Chat-7B model, it always exits without any reason and there are no error prompts
#1090
anyiz
closed
5 months ago
2
[BUG] finetune.py中的实现,默认会跳过用户设置的system_message,而使用默认的"You are a helpful assistant."
#1089
yetionyo
closed
7 months ago
2
微调Qwen-7B-Chat模型后量化模型无故退出程序也不报错gpu也没有oom
#1088
anyiz
closed
6 months ago
1
[BUG] 莫名其妙的TypeError: 'NoneType' object is not iterable
#1087
ElinLiu0
closed
7 months ago
3
Qwen對繁體中文的識別及生成能力
#1086
ACBBZ
closed
6 months ago
2
使用qwen模型进行RLHF时出错
#1085
128Ghe980
closed
6 months ago
2
[BUG] 回复 hovering hovering hovering hovering
#1083
aaabbb8853
closed
5 months ago
1
微调后的模型,用vllm推理出来的结果都是空是什么原因呀
#1082
lalalabobobo
closed
5 months ago
3
[BUG] You can't train a model that has been loaded with `device_map='auto'` in any distributed mode
#1081
jiangliqin
closed
7 months ago
1
Qlora训练完部署具体要进行哪些细节
#1079
xtanitfy
closed
7 months ago
3
[BUG] <title> 两个显卡训练,保存模型时出错
#1078
xtanitfy
closed
7 months ago
3
windows部署,启动时报错Tokenizer class Qwen2Tokenizer does not exist or is not currently imported
#1077
logan4753
closed
7 months ago
5
关于Chat模型数据集
#1076
ftgreat
closed
7 months ago
1
[BUG] lora微调到1000save_steps保存模型报错,提示文件不存在
#1075
jiangliqin
closed
5 months ago
6
请问一下,qwen 的vllm部署和原始的部署出来的结果存在不一致,结果是经过多次反复验证的。请问有没有解决的方案
#1074
qingjiaozyn
closed
6 months ago
4
6张A40卡并行微调Qwen-14B-chat,推理时报错
#1073
HooRin
closed
5 months ago
4
请问通义千问1.0的bash finetune/finetune_qlora_single_gpu.sh微调命令,对通义千问1.5适用吗?
#1072
annian101
closed
7 months ago
2
vLLM是不是支持量化模型了?
#1070
rabum
closed
7 months ago
1
请问模型的并发能力
#1067
rabum
closed
7 months ago
1
qwen大模型在基于背景知识总结答案过程中,针对特殊符号比如竖线、横杠之类的会存在丢失情况
#1066
WangxuP
closed
7 months ago
4
lora微调,训练集长度扩展问题
#1065
fanbooo
closed
7 months ago
4
Update evaluate_plugin.py
#1063
seanxuu
closed
6 months ago
0
[BUG] 用anything-llm链接openai api格式的本地API部署出现问题
#1062
afezeriaWrnbbmm
closed
7 months ago
3
我微调了预训练模型Qwen-7B如何让他实现多轮问答产生上下文联系
#1061
anyiz
closed
7 months ago
1
[BUG] openai_api request with functions payload returns 400
#1057
akinlong
closed
7 months ago
1
关于MMLU的测试结果的疑问
#1052
andeyeluguo
closed
7 months ago
4
[BUG] <title>为什么千问14B的config里面的seq_length 和max_position_embedding不一样,而7B和72B是一样的。同时vllm的千问的实现中没有Q*logn的实现
#1051
wqh17101
closed
7 months ago
3
流式推理时,完成一轮对话后,模型输出最后一个字后,有没有什么结束标识呢?
#1049
stevin-dong
closed
7 months ago
6
流式推理时,完成一轮对话后,模型输出最后一个字后,有没有什么结束标识呢?
#1048
stevin-dong
closed
8 months ago
0
[BUG] <关于generate阶段是否支持embedding输入的问题>
#1047
AZYoung233
closed
7 months ago
6
[BUG] <title>Value Error: Tokenizer class QwenTokenizier does not exist or is not currently imported
#1046
h66840
closed
8 months ago
1
[BUG] <title>微调qwen-chat-7B-int4的时候Target module QuantLinear() is not supported.
#1045
Yining0907
closed
8 months ago
4
请看下这个问题是什么原因:Token indices sequence length is longer than the specified maximum sequence length for this model (579376 > 512). Running this sequence through the model will result in indexing errors
#1044
chesp
closed
6 months ago
3
[BUG] <title>单机8卡A100进行Qwen-72B-chat-Int4 QLora训练时 出现OOM报错
#1043
KevinFan0
closed
4 months ago
7
在chat上SFT max_seq_len开大后指令遵循能力下降很严重
#1042
menghonghan
closed
6 months ago
2
Add peft note
#1041
jklj077
closed
8 months ago
0
[BUG] <qwen-7b-chat-PRO 全量SFT max_seq_len开到32k相比2048 对话&理解能力下降非常严重>
#1040
menghonghan
closed
8 months ago
0
[BUG]使用model.chat()进行infer时报错
#1039
128Ghe980
closed
8 months ago
0
update openai_api: support stop words for streaming chat
#1038
tuhahaha
closed
8 months ago
0
计算推理速度的profile.py不能运行
#1037
rabum
closed
7 months ago
2
qlora微调后无法读取权重文件
#1036
zzzcccxx
closed
8 months ago
5
使用QwenLM/vllm-gptq运行Qwen-14B-Chat-Int4报错:ValueError: The input size is not aligned with the quantized weight shape. This can be caused by too large tensor parallel size.
#1035
huangyunxin
closed
6 months ago
2
Flash attention import失败
#1034
rabum
closed
8 months ago
3
add trt docker file
#1033
haiasd
closed
8 months ago
0
Update README.md
#1032
logicwong
closed
8 months ago
0
Previous
Next