Qwen3-VL-30B-A3B-Instruct qwen3-vl-30b-a3b-instruct

qwen · qwen vision efficient

#35 overall

params
30.5B (A3B)
arch
moe
context
256k
license
apache-2.0 open weights
released
Sep 2025
downloads/30d
646.1k

# why ranked

Overall: #35 — score 18.2 ● medium — one signal missing, weights renormalized

signalweightinput (0–100)
aa_intelligence 0.50 10.0
arena_elo 0.30
bench_composite 0.20 38.7

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 10.0 Artificial Analysis
AA Math Index 72.3 Artificial Analysis
Ai2d 0.8 / 1 LLM Stats
AIME 2025 0.7 / 1 LLM Stats
Arena-hard-v2 0.6 / 1 LLM Stats
Bfcl-v3 0.7 / 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 0.9 / 1 LLM Stats
Mm-mt-bench 8.1 LLM Stats

# where to run

providerquantctx$/M in$/M out$/M blendtpsuptime
Alibabaunknown 128k $0.13$0.52$0.23 100.0%
DeepInfrafp8 256k $0.15$0.60$0.26 100.0%
Novitabf16 128k $0.20$0.70$0.33
Phalaunknown 125k $0.20$0.70$0.33 97.9%
SiliconFlowfp8 256k $0.29$1.00$0.47 100.0%

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

# variants

Qwen3-VL-30B-A3B-Thinking

source aliases
aa
qwen3-vl-30b-a3b-instruct
llmstats
qwen3-vl-30b-a3b-instruct
openrouter
qwen/qwen3-vl-30b-a3b-instruct