Qwen/QwQ-32B · Discussions

Test using M1 Max (64G) and Word

#75 opened 5 days ago by

gptlocalhost

QWQ-32B在私有部署的时候是否可以规避掉审查机制？

#74 opened 6 days ago by

youlchats

请问qwq的知识截止日期是什么时候呀？

#73 opened 16 days ago by

hexiaoyu

-

#72 opened 24 days ago by

Amyww

Ifeval score 55 on inst-level loose-accuracy.

#70 opened 27 days ago by

Jamesunnc

Upload hokm96.xlsx

#69 opened 27 days ago by

ashkanpourali

poor south west Asia languages

1

#68 opened 29 days ago by

rastegar

Sliding Window Attention

#67 opened 30 days ago by

cferggie

Update README.md

#66 opened about 1 month ago by

sujiij

使用react agent解决复杂问题时只会调用一次函数

1

#64 opened about 1 month ago by

tyler1990

reasoning dead loop.

#63 opened about 1 month ago by

creampx

Seeking Advice on Fine-tuning QWQ-32B Model

2

#62 opened about 1 month ago by

aaditya

Steps to deploy a production ready service for QwQ on AWS using serverless GPUs

7

#61 opened about 1 month ago by

samagra14

Etude

#60 opened about 1 month ago by

Nino-ogvng

Upload gen_ai(proj_47).ipynb

#59 opened about 1 month ago by

harshith1411

How does LiveCodeBench test?

3

#58 opened about 1 month ago by

cizhenshi

Create test

2

#57 opened about 1 month ago by

Amyww

Upload 99ed5d4766696bd4ebc26e5d9c23e982.png

1

#56 opened about 1 month ago by

axingd

Best practice for QwQ-32B evaluation

5

3

#55 opened about 2 months ago by

wangxingjun778

Create test.txt

#54 opened about 2 months ago by

xxxx443117

How to continue pt / sft on this model，any suggestions?

4

#53 opened about 2 months ago by

Ken0102030405

Infinite repetitive thinking for this case:

1

#52 opened about 2 months ago by

zhaocc1106

Is QwQ-32B used for custom NER extraction? Does this model work better compared to other open-source BERT transformers?

1

#51 opened about 2 months ago by

amiirhmza

Budget forcing?

1

#50 opened about 2 months ago by

mwettach

Allow prefilling assistant message

1

#49 opened about 2 months ago by

tomasmcm

Update README.md

#48 opened about 2 months ago by

Bschleter

Day of the week

2

#47 opened about 2 months ago by

jac-jim

Intermittent CUDA error with model.generate() using device_map="auto" and 3 GPUs

#46 opened about 2 months ago by

lucmaz98

Create Call Center Tunaiku 0818836245

#45 opened about 2 months ago by

Jokiio

Does Macbook M1 max 64GB run this model well?

1

#44 opened about 2 months ago by

mrk83

Too many "cross-validate" and "another method"

4

2

#43 opened about 2 months ago by

AaronFeng753

RuntimeError: Error(s) in loading state_dict for Qwen2ForCausalLM:

1

#42 opened about 2 months ago by

XuehangCang

8GB GPU can run this,10t/s

2

#41 opened about 2 months ago by

wqerrewetw

When answering questions in Chinese, the model frequently terminates prematurely (outputs the end token). Is this a common problem?

1

#40 opened about 2 months ago by

zhangw355

Refining QWQ Model Output: Direct Responses Without Step-by-Step Reasoning

6

1

#39 opened about 2 months ago by

gslinx

It's challenging for QwQ to generate long codes...

2

#38 opened about 2 months ago by

DXBTR74

Nice work... Cant-believe-its-just-32B-performance even with various different tones system prompt.

#37 opened about 2 months ago by

imoc

function call时有办法跳过think吗？

1

2

#36 opened about 2 months ago by

zhaocc1106

Failed to parse Jinja template:

2

#35 opened about 2 months ago by

Vicnent

Obligatory question about model sizes...

#34 opened about 2 months ago by

MrDevolver

This model beats Qwen Max!

1

7

#33 opened about 2 months ago by

MrDevolver

remove part about long context modifications

#32 opened about 2 months ago by

nbroad

add a reasoning effort option

2

#31 opened about 2 months ago by

TheBigBlockPC

用vllm时应该是什么参数

6

#30 opened about 2 months ago by

daiwk

遇到复杂问题时，开始推理时有<think>，推理结束了还没有</think>

6

#29 opened about 2 months ago by

digits12

Is this model native 128K context length, or YaRN extended?

7

#28 opened about 2 months ago by

danielhanchen

docs: update README.md

#27 opened about 2 months ago by

eltociear

Thanks a lot for sharing this model!

5

#26 opened about 2 months ago by

FalconNet

Doesn't Generate `<think>` tags

3

#25 opened about 2 months ago by

bingw5

【乱码问题】输入约1w长度，输出超过1000长度，结尾部分会乱码。稳定复现！

1

#24 opened about 2 months ago by

chizhu