Qwen

Qwen2.5 VL 72B Instruct

by Qwentext+image->text1 endpoint
Neura Intelligence Index
18.0/ 100
Rank
#297
Confidence
medium
Benchmarks
4(10% coverage)

Overview

Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.

Capabilities

Text generation
Image understanding
Structured output
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
TextImage
Output
Text

Technical Specifications

Context Window
128K tokens
Max Output
115.2K tokens
Knowledge Cutoff
2024-06-30
Tokenizer
Qwen
Uptime (24h)
10000.0%
Parameters
72000000000B
License
tongyi_qianwenOpen Weights
Country
🇨🇳CN
Release Date
January 26, 20251y

Supported Parameters

frequency_penaltylogit_biaslogprobsmax_tokenspresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetop_ktop_logprobstop_p

Benchmark Performance

multimodal

BenchmarkScoreSource
MMMU
70.2
verified
MMMU-Pro
51.1
verified
ScreenSpot Pro
43.6
verified

agent

BenchmarkScoreSource
OSWorld
8.8
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.80
Output$1.00
Cache read$0.40
Added
February 1, 2025
Last synced 9/16/2026