Qwen/QwQ-32B-Preview · Discussions

question about output length

1

#59 opened 30 days ago by

skylerr

Remove CoT prompt from chat template

1

#58 opened about 1 month ago by

SumanthRH

Why is this not in inference api?

#56 opened about 1 month ago by

Reality123b

请问你们qwq的prompt是什么？推理有什么特殊设置吗？我这边复现live-codebench只有35～38分。（live-code-bench-0801-1101）

#53 opened about 2 months ago by

xxzcc

Does QwQ support function calling?

3

#52 opened about 2 months ago by

nanma

Tried the "strawberry" demo case but got wrong answer😂

2

#51 opened about 2 months ago by

HuggingLianWang

🖐️🖐️ Can we please get a 7B reasoning model? 😎🤗🤗😍

1

#50 opened about 2 months ago by

so-anyway

Request: DOI

#49 opened 2 months ago by

daniqt2

Can't reproduce the evaluation result of GPQA dataset

5

#47 opened 2 months ago by

Rinn000

Adding Evaluation Results

#44 opened 2 months ago by

QueCumber

Evaluation Details of QwQ-32B-Preview on LiveCodeBench Benchmark

#43 opened 3 months ago by

survivi

Independent Evaluation of QwQ-32B-Preview on MMLU-Pro Benchmark

1

#42 opened 3 months ago by

yaronr

Can't copy the full reasoning

2

#41 opened 3 months ago by

Hamora

Create bsnwritingservices

#40 opened 3 months ago by

carlo34

https://huggingface.co/Qwen/QwQ-32B-Preview/resolve/main/.gitattributes

#39 opened 3 months ago by

tongtong22

Floating point exception

2

#38 opened 3 months ago by

zhongqy

My prompt induced infinite loops with QwQ

1

#35 opened 3 months ago by

hanaoverride

docs: update README.md

#34 opened 3 months ago by

eltociear

Which one is more risky, type 1 or type 2?

2

#33 opened 3 months ago by

Hanialy

Silly tavern template?

4

#32 opened 3 months ago by

BigBeavis

Update README.md

#31 opened 3 months ago by

cumanpc

Adding Evaluation Results

#30 opened 3 months ago by

leaderboard-pr-bot

Fine-tuning or system prompt for fiction writing?

1

#28 opened 3 months ago by

advokat

Kkk

#27 opened 3 months ago by

Iksksks

Update README.md

#26 opened 3 months ago by

OMSRI-SRI-RADHAKRSNA-1728

Will the training code and data be released?

1

#25 opened 3 months ago by

WbjuSrceu

early exit reasoning inference

#24 opened 3 months ago by

miniTsl

RAG architecture based on the model family qwen2.5

#22 opened 3 months ago by

shevel

This is way too much... USB? Yes. U SB.

3

#21 opened 3 months ago by

imoc

该模型只适合聊天，没有其他能力

2

#20 opened 3 months ago by

royabcde

trick for less token usage and less halucination

4

#19 opened 3 months ago by

gopi87

multi GPU inferencing

2

#18 opened 3 months ago by

cjj2003

Install QwQ 32B Locally - Step by Step Video Tutorial

#17 opened 3 months ago by

fahdmirzac

Adds Chinese characters to responses

7

#16 opened 3 months ago by

maxbenk

Problem with specific output format

#15 opened 3 months ago by

sszymczyk

UwU

#14 opened 3 months ago by

PotatoOff

About Releasing Some Examples of QWQ's Training Dataset

#13 opened 3 months ago by

tanhuajie2001

Nice name QAQ. I'll later upload a 4.7bpw quantized model if it works.

#12 opened 3 months ago by

imoc

QuietSTAR?

1

#9 opened 3 months ago by

albatrossbirdie

Inference Speed Benchmark and GPU memeory usage

#8 opened 3 months ago by

Yunxz

Some Badcases

1

#7 opened 3 months ago by

yirgacheffe

ways to contain the thought process from users for a cleaner interface

#5 opened 3 months ago by

Steelskull

A huge thanks!

#2 opened 3 months ago by

owao

ggggggggg

2

#1 opened 3 months ago by

Smougueni

question about output length

Remove CoT prompt from chat template

Why is this not in inference api?

请问你们qwq的prompt是什么？ 推理有什么特殊设置吗？我这边复现live-codebench只有35～38分。（live-code-bench-0801-1101）

Does QwQ support function calling?

Tried the "strawberry" demo case but got wrong answer😂

🖐️🖐️ Can we please get a 7B reasoning model? 😎🤗🤗😍

Request: DOI

Can't reproduce the evaluation result of GPQA dataset

Adding Evaluation Results

Evaluation Details of QwQ-32B-Preview on LiveCodeBench Benchmark

Independent Evaluation of QwQ-32B-Preview on MMLU-Pro Benchmark

Can't copy the full reasoning

Create bsnwritingservices

https://huggingface.co/Qwen/QwQ-32B-Preview/resolve/main/.gitattributes

Floating point exception

My prompt induced infinite loops with QwQ

docs: update README.md

Which one is more risky, type 1 or type 2?

Silly tavern template?

Update README.md

Adding Evaluation Results

Fine-tuning or system prompt for fiction writing?

Kkk

Update README.md

Will the training code and data be released?

early exit reasoning inference

RAG architecture based on the model family qwen2.5

This is way too much... USB? Yes. U SB.

该模型只适合聊天，没有其他能力

trick for less token usage and less halucination

multi GPU inferencing

Install QwQ 32B Locally - Step by Step Video Tutorial

Adds Chinese characters to responses

Problem with specific output format

UwU

About Releasing Some Examples of QWQ's Training Dataset

Nice name QAQ. I'll later upload a 4.7bpw quantized model if it works.

QuietSTAR?

Inference Speed Benchmark and GPU memeory usage

Some Badcases

ways to contain the thought process from users for a cleaner interface

A huge thanks!

ggggggggg

请问你们qwq的prompt是什么？推理有什么特殊设置吗？我这边复现live-codebench只有35～38分。（live-code-bench-0801-1101）