Back to Explore
Qwen2.5 VL 32B Instruct
Alibaba·Released 2025-03-21
Open Source32B128K ctxApache 2.0Multimodal
About
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...
Related Models
Pricing
Input
$0.20
per 1M tokens
Output
$0.60
per 1M tokens
Blended
$0.30
per 1M tokens
Cheaper than 65% of models. Median price is $0.67/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.30
Monthly
$9.00
vs. Similar Models
Seed-OSS-36B-Instruct
$0.300%
NVIDIA Nemotron Nano 12B v2 VL (Reasoning)
$0.300%
Olmo 3.1 32B Instruct
$0.300%
Mistral: Saba
$0.300%
Performance
Context Window
128K
tokens
Larger than 16% of models
Context Window Comparison
Inception: Mercury 2
128KSame
Perplexity: Sonar Reasoning Pro
128KSame
Upstage: Solar Pro 3
128KSame
Open Source
apache-2.032BGGUF / GPTQ / AWQ
Downloads
424.1K
Likes
491
VRAM (FP16)
24-48 GB
GPU
A6000 / M3 Ultra