Qwen: QwQ 32B

Text input Text output Free Option
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, especially hard problems. QwQ-32B is the medium-sized reasoning model, which is capable of achieving competitive performance against state-of-the-art reasoning models, e.g., DeepSeek-R1, o1-mini.

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

This model supports the following parameters:

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

This model supports the following features:

Reasoning Response Format
Performance Summary

Qwen: QwQ 32B, a reasoning model from Qwen, demonstrates exceptional overall performance, particularly excelling in speed and reliability. It consistently ranks among the fastest models, achieving an Infinityth percentile in speed across seven benchmarks. Pricing is competitive, placing it in the 54th percentile. Reliability is a significant strength, with a perfect 100% success rate across all benchmarks, indicating minimal technical failures. In terms of specific performance, QwQ 32B shows strong capabilities in reasoning and knowledge-based tasks, achieving 98.0% accuracy in Reasoning (95th percentile) and 99.0% in General Knowledge (71st percentile). Its Ethics performance is perfect at 100% accuracy, notably being the most accurate model at its price point and speed. Coding accuracy is also robust at 91.0% (80th percentile). While its initial Instruction Following (Baseline) result was 0.0%, a subsequent run showed a more representative 53.0% accuracy. The model's primary strength lies in its reasoning capabilities, aligning with its description as a reasoning model. Its main weakness appears to be inconsistency in instruction following, as evidenced by the disparate results.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.075
Completion $0.15

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
DeepInfra
DeepInfra | qwen/qwq-32b 131K $0.075 / 1M tokens $0.15 / 1M tokens
Nebius
Nebius | qwen/qwq-32b 131K $0.5 / 1M tokens $1.5 / 1M tokens
InferenceNet
InferenceNet | qwen/qwq-32b 16K $0.075 / 1M tokens $0.15 / 1M tokens
Groq
Groq | qwen/qwq-32b 131K $0.075 / 1M tokens $0.15 / 1M tokens
Hyperbolic
Hyperbolic | qwen/qwq-32b 131K $0.4 / 1M tokens $0.4 / 1M tokens
SambaNova
SambaNova | qwen/qwq-32b 16K $0.075 / 1M tokens $0.15 / 1M tokens
Cent-ML
Cent-ML | qwen/qwq-32b 40K $0.075 / 1M tokens $0.15 / 1M tokens
Fireworks
Fireworks | qwen/qwq-32b 131K $0.075 / 1M tokens $0.15 / 1M tokens
Together
Together | qwen/qwq-32b 131K $1.2 / 1M tokens $1.2 / 1M tokens
Nineteen
Nineteen | qwen/qwq-32b 40K $0.075 / 1M tokens $0.15 / 1M tokens
NextBit
NextBit | qwen/qwq-32b 32K $0.15 / 1M tokens $0.4 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by qwen