Qwen: QwQ 32B

Text input Text output
Author's Description

QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tuned models, QwQ, which is capable of thinking and reasoning, can achieve significantly enhanced performance in downstream tasks,...

Key Specifications
Cost
$$$
Context
131K
Parameters
32B
Released
Mar 05, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Seed Frequency Penalty Top P Min P Response Format Reasoning Temperature Stop Presence Penalty Include Reasoning Max Tokens
Features

This model supports the following features:

Response Format Reasoning
Performance Summary

Qwen: QwQ 32B, a medium-sized reasoning model from the Qwen series, demonstrates exceptional speed, consistently ranking among the fastest models with an Infinityth percentile across nine benchmarks. Its pricing is competitive, placing it in the 52nd percentile across eight benchmarks. Furthermore, QwQ 32B exhibits outstanding reliability, boasting a 99% success rate across nine benchmarks, indicating minimal technical failures. In terms of performance across categories, QwQ 32B shows significant strengths in complex reasoning tasks, achieving 98.0% accuracy (93rd percentile) in the Reasoning benchmark. It also excels in Ethics, reaching a perfect 100.0% accuracy, making it the most accurate model at its price point and among models of comparable speed. Mathematics is another strong area, with 95.9% accuracy (91st percentile). General Knowledge and Email Classification also show robust performance at 99.0% and 98.0% accuracy respectively. While its Coding accuracy is respectable at 91.0% (70th percentile), a notable weakness is observed in Hallucinations, where it achieved 86.0% accuracy (31st percentile), suggesting room for improvement in acknowledging uncertainty. Instruction Following presents a mixed picture, with one benchmark showing 53.0% accuracy and another indicating 0.0%, which warrants further investigation. Overall, QwQ 32B is a highly reliable and fast model with strong reasoning and ethical capabilities.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.15
Completion $0.58

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
DeepInfra
DeepInfra | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
Nebius
Nebius | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
InferenceNet
InferenceNet | qwen/qwq-32b 16K $0.15 / 1M tokens $0.58 / 1M tokens
Groq
Groq | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
Hyperbolic
Hyperbolic | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
SambaNova
SambaNova | qwen/qwq-32b 16K $0.15 / 1M tokens $0.58 / 1M tokens
Cent-ML
Cent-ML | qwen/qwq-32b 40K $0.15 / 1M tokens $0.58 / 1M tokens
Fireworks
Fireworks | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
Together
Together | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
Nineteen
Nineteen | qwen/qwq-32b 40K $0.15 / 1M tokens $0.58 / 1M tokens
NextBit
NextBit | qwen/qwq-32b 32K $0.15 / 1M tokens $0.58 / 1M tokens
SiliconFlow
SiliconFlow | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
Nebius
Nebius | qwen/qwq-32b 131K $0.15 / 1M tokens $0.58 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by qwen