AIbase
EN
Home
AI Models Finder
Qwq 32B Preview
Q

Qwq 32B Preview

An experimental research model focused on enhancing the inference ability of artificial intelligence, particularly excelling in mathematics and programming. It has the ability of in - depth self - reflection and self - questioning, but there are some limitations in language mixing and recursive reasoning patterns.
Intelligence(Medium)
Speed(Relatively Slow)
Input Supported Modalities
No
Is Reasoning Model
32,768
Context Window
32,768
Maximum Output Tokens
2024-11-28
Knowledge Cutoff
Go Compare

Pricing

￥1.08 /M tokens
Input
￥1.44 /M tokens
Output
￥4.81 /M tokens
Blended Price

Quick Simple Comparison

Input

Output

Qwen2.5 Turbo
Qwen Turbo
Wan2.2

Basic Parameters

QwQ 32B-PreviewTechnical Parameters
Parameter Count
32,500.0M
Context Length
32.77k tokens
Training Data Cutoff
2024-11-28
Open Source Category
Open Weights (Permissive License)
Multimodal Support
Text Only
Throughput
99
Release Date
2024-11-27
Response Speed
51.34,079 tokens/s

Benchmark Scores

Below is the performance of QwQ 32B-Preview in various standard benchmark tests. These tests evaluate the model's capabilities in different tasks and domains.
Intelligence Index
42.6
Large Language Model Intelligence Level
Coding Index
18.74
Indicator of AI model performance on coding tasks
Math Index
68.17
Capability indicator in solving mathematical problems, mathematical reasoning, or performing math-related tasks
MMLU Pro
64.8
Massive Multitask Multimodal Understanding - Testing understanding of text, images, audio, and video
GPQA
55.7
Graduate Physics Questions Assessment - Testing advanced physics knowledge with diamond science-level questions
HLE
4.8
The model's comprehensive average score on the Hugging Face Open LLM Leaderboard
LiveCodeBench
33.7
Specific evaluation focused on assessing large language models' ability in real-world code writing and solving programming competition problems
SciCode
3.8
The model's capability in code generation for scientific computing or specific scientific domains
HumanEval
86.7
Score achieved by the AI model on the specific HumanEval benchmark test set
Math 500 Score
91
Score on the first 500 larger, more well-known mathematical benchmark tests
AIME Score
45.3
An indicator measuring an AI model's ability to solve high-difficulty mathematical competition problems (specifically AIME level)
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase