Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · Qwen3-VL-8B-InstructQwen14 results · 13 benchmarks
Model card

Qwen3-VL-8B-Instruct.

QwenLarge language model

Added from Papers with Code MMLU-Pro refresh on 2026-05-19.

§ 02 · Benchmarks

Every benchmark Qwen3-VL-8B-Instruct has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01DocVQAComputer Vision · Document Understandinganls96.1%#7/21source ↗
02MMStarMultimodal · Image-Text-to-Textaccuracy70.9%#13/21source ↗
03RealWorldQAMultimodal · Visual Question Answeringaccuracy71.5%#14/23source ↗
04MMMUMultimodal · Image-Text-to-Textaccuracy69.6%#16/36source ↗
05MVBenchMultimodal · Video Understandingaccuracy68.7%#16/20source ↗
06OSWorldAgentic AI · Web & Desktop Agentssuccess-rate33.9%#16/28source ↗
07Video-MMEMultimodal · Video Understandingaccuracy71.4%#17/24source ↗
08AIME 2025Reasoning · Mathematical Reasoningaccuracy45.9%#21/22source ↗
09MMMU-ProMultimodal · Visual Question Answeringaccuracy55.9%#21/31source ↗
10LiveCodeBenchComputer Code · Code Generationpass-139.3%#23/24source ↗
11MMLUReasoning · Commonsense Reasoningaccuracy80.7%#51/64source ↗
12MMLU-ProReasoning · Commonsense Reasoningaccuracy71.6%#59/732025-11-26source ↗
13MMLU-ProReasoning · Commonsense Reasoningaccuracy71.6%#59/73source ↗
14GPQA DiamondReasoning · Multi-step Reasoningaccuracy34.7%#74/74source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Qwen3-VL-8B-Instruct actually performs.

Computer Vision
1
benchmark
avg rank #7.0
Agentic AI
1
benchmark
avg rank #16.0
Multimodal
6
benchmarks
avg rank #16.2
Computer Code
1
benchmark
avg rank #23.0
Reasoning
4
benchmarks
avg rank #52.8
§ 04 · Papers

1 paper with results for Qwen3-VL-8B-Instruct.

  1. 2025-11-26· 13 results

    Qwen3-VL Technical Report

§ 05 · Related models

Other Qwen models scored on Codesota.

Qwen3-VL-235B-A22B-Instruct
9 results · 1 SOTA
Qwen3-VL-235B-A22B-Thinking
9 results
Qwen3-Coder-Next
7 results
Qwen2.5-Plus
1 result
Qwen2.5-VL-72B
1 result
Qwen3-235B-A22B (Base)
1 result
Qwen3.5-122B-A10B
1 result
Qwen3.5-27B
1 result
§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
13
results
paperswithcode
1
result
0 of 14 rows marked verified.