Qwen

Qwen3 VL 8B Instruct

by Qwentext+image->text2 endpoints
Neura Intelligence Index
12.3/ 100
Rank
#317
Confidence
medium
Benchmarks
5(13% coverage)

Overview

Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon...

Capabilities

Text generation
Image understanding
Tool use / Function calling
Structured output
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
ImageText
Output
Text

Technical Specifications

Context Window
262.1K tokens
Max Output
32.8K tokens
Tokenizer
Qwen3
Uptime (24h)
9998.4%
Parameters
9000000000B
License
apache_2_0Open Weights
Country
🇨🇳CN
Release Date
September 22, 202512mo

Supported Parameters

frequency_penaltylogit_biaslogprobsmax_tokenspresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

math

BenchmarkScoreSource
AIME 2025
45.9
verified

multimodal

BenchmarkScoreSource
MMMU-Pro
55.9
verified
ScreenSpot Pro
54.6
verified
CharXiv Reasoning
46.4
verified

agent

BenchmarkScoreSource
OSWorld
33.9
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.12
Output$0.46
Added
October 14, 2025
Last synced 9/16/2026