Qwen3-VL-8B-Thinking qwen3-vl-8b-thinking

qwen · qwen vision reasoning efficient

#39 overall

params
8.8B
arch
dense
context
250k
license
apache-2.0 open weights
released
Oct 2025
reasoning
yes
downloads/30d
102.3k

# why ranked

Overall: #39 — score 15.7 ● medium — one signal missing, weights renormalized

signalweightinput (0–100)
aa_intelligence 0.50 10.6
arena_elo 0.30
bench_composite 0.20 28.4

missing signals are dropped and the remaining weights renormalized — never imputed.

full methodology

# trend

collecting history — trends appear after a few daily runs.

# benchmarks

benchmarkscoresourcedate
AA Intelligence Index 10.6 Artificial Analysis
AA Math Index 30.7 Artificial Analysis
Ai2d 0.8 / 1 LLM Stats
AIME 2025 0.8 / 1 LLM Stats
Arena-hard-v2 0.5 / 1 LLM Stats
Bfcl-v3 0.6 / 1 LLM Stats
Blink 0.7 / 1 LLM Stats
Cc-ocr 0.8 / 1 LLM Stats
Charadessta 0.6 / 1 LLM Stats
Charxiv-d 0.9 / 1 LLM Stats
Charxiv-r 0.5 / 1 LLM Stats
Creative-writing-v3 0.8 / 1 LLM Stats
Docvqatest 1.0 / 1 LLM Stats
Mm-mt-bench 8.0 LLM Stats

# where to run

providerquantctx$/M in$/M out$/M blendtpsuptime
Alibabaunknown 128k $0.12$1.36$0.43

blended = (3·input + output) / 4 per 1M tokens · cheapest first

source aliases
aa
qwen3-vl-8b-reasoning
llmstats
qwen3-vl-8b-thinking
openrouter
qwen/qwen3-vl-8b-thinking