Qwen: Qwen2.5 VL 32B Instruct

Text input Image input Text output
Author's Description

Qwen2.5-VL-32B is a multimodal vision-language model fine-tuned through reinforcement learning for enhanced mathematical reasoning, structured outputs, and visual problem-solving capabilities. It excels at visual analysis tasks, including object recognition, textual...

Key Specifications
Cost
$$$
Context
128K
Parameters
32B
Released
Mar 24, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Frequency Penalty Structured Outputs Top P Logprobs Response Format Temperature Stop Presence Penalty Max Tokens Logit Bias Top Logprobs
Features

This model supports the following features:

Structured Outputs Response Format
Performance Summary

Qwen2.5 VL 32B Instruct demonstrates competitive response times, performing among the faster models with a 59th percentile speed ranking across benchmarks. It also offers competitive pricing, ranking in the 52nd percentile. Notably, the model exhibits exceptional reliability, achieving a 95% success rate, indicating consistent and usable responses. In terms of performance across categories, Qwen2.5 VL 32B Instruct shows remarkable strength in Ethics, achieving perfect 100% accuracy, making it the most accurate model at its price point and among models of similar speed. It also performs well in General Knowledge and Email Classification with 98.0% accuracy in both, placing it around the 50th and 60th percentile respectively. Its Coding performance is solid at 84.0% accuracy. However, the model exhibits notable weaknesses in Instruction Following and Reasoning, with accuracies of 40.4% and 58.3% respectively. The Reasoning benchmark, in particular, shows a very long duration, placing it in the 2nd percentile for speed in that category. Overall, Qwen2.5 VL 32B Instruct excels in visual analysis and ethical understanding, while its reasoning and complex instruction following capabilities could benefit from further enhancement.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.2
Completion $0.6

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Fireworks
Fireworks | qwen/qwen2.5-vl-32b-instruct 128K $0.2 / 1M tokens $0.6 / 1M tokens
DeepInfra
DeepInfra | qwen/qwen2.5-vl-32b-instruct 128K $0.2 / 1M tokens $0.6 / 1M tokens
Chutes
Chutes | qwen/qwen2.5-vl-32b-instruct 16K $0.2 / 1M tokens $0.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by qwen