llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsQwen3 30B A3B

Alibaba Cloud / Qwen Team model product

Qwen3 30B A3B

Qwen3-30B-A3B is a smaller Mixture-of-Experts (MoE) model from the Qwen3 series by Alibaba, with 30.5 billion total parameters and 3.3 billion activated parameters. Features hybrid thinking/non-thinking modes, support for 119 languages, and enhanced agent capabilities. It aims to outperform previous models like QwQ-32B while using significantly fewer activated parameters.

Updated Aug 10, 2026. Default version: Qwen3 30B A3B

Compare
LLMBoard score39.7Qwen3 30B A3B
Coverage60%5 benchmark families
Context window128KTokens
Official input priceN/AOfficial price unavailable

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
Qwen3 30B A3B
Released
Apr 29, 2025
Knowledge cutoff
Unknown
Parameters
30.5B
Context window
128K
Max output
128K
Inputs
text
Outputs
text
Open weights
No
License
Apache 2.0

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

Qwen3 30B A3B category scores

Benchmark results

Published benchmark records for the scored version Qwen3 30B A3B.

8 rows
Columns

Show columns

Arena Hard0.932692.0%CAug 7, 2026
BFCL0.771140.0%CAug 7, 2026
Multi-IF0.7132036.8%CAug 7, 2026
LiveBench0.7183854.0%CAug 7, 2026
AIME 20240.8275350.0%CAug 7, 2026
LiveCodeBench0.6317358.3%CAug 7, 2026
AIME 20250.78711423.9%CAug 7, 2026
GPQA0.714623337.5%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

58 rows
Columns

Show columns

textjapanese1571251.9665N/AAug 6, 2026
textfrench1611353.3322N/AAug 6, 2026
textkorean1681262.6439N/AAug 6, 2026
text style controlfrench1691364.6322N/AAug 6, 2026
textindustry mathematical1711364.51,577N/AAug 6, 2026
textgerman1721308.3727N/AAug 6, 2026
text style controljapanese1721251.5665N/AAug 6, 2026
textspanish1741318.6410N/AAug 6, 2026
text style controlindustry mathematical1751367.11,577N/AAug 6, 2026
textpolish1761299.91,956N/AAug 6, 2026
textchinese1811355.51,252N/AAug 6, 2026
textmath1831355.11,701N/AAug 6, 2026
text style controlkorean1841268.1439N/AAug 6, 2026
text style controlmath1861352.61,701N/AAug 6, 2026
text style controlpolish1881303.91,956N/AAug 6, 2026
text style controlspanish1881324.5410N/AAug 6, 2026
text style controlchinese1891370.01,252N/AAug 6, 2026
text style controlgerman1901304.1727N/AAug 6, 2026
textexpert1931313.91,313N/AAug 6, 2026
text style controlexpert1951343.01,313N/AAug 6, 2026
text style controlcoding1991386.54,526N/AAug 6, 2026
textcoding2021338.14,526N/AAug 6, 2026
textindustry business and management and financial operations2021308.84,010N/AAug 6, 2026
textindustry medicine and healthcare2021317.91,621N/AAug 6, 2026
textindustry software and it services2021340.58,301N/AAug 6, 2026
textlonger query2051313.34,446N/AAug 6, 2026
textnon english2051294.813,064N/AAug 6, 2026
texthard prompts2061315.510,682N/AAug 6, 2026
texthard prompts english2071331.45,806N/AAug 6, 2026
textindustry legal and government2071313.71,770N/AAug 6, 2026
textindustry life and physical and social science2071321.14,606N/AAug 6, 2026
text style controlindustry business and management and financial operations2071330.04,010N/AAug 6, 2026
text style controlindustry software and it services2071371.48,301N/AAug 6, 2026
textenglish2081335.313,375N/AAug 6, 2026
textrussian2081288.81,898N/AAug 6, 2026
text style controlhard prompts english2081362.35,806N/AAug 6, 2026
text style controllonger query2081339.54,446N/AAug 6, 2026
textoverall2091317.226,448N/AAug 6, 2026
textexclude ties2101260.818,327N/AAug 6, 2026
textmulti turn2101307.54,514N/AAug 6, 2026
text style controlhard prompts2111345.410,682N/AAug 6, 2026
textindustry writing and literature and language2131291.15,958N/AAug 6, 2026
textinstruction following2141285.76,120N/AAug 6, 2026
textcreative writing2191272.23,361N/AAug 6, 2026
text style controlenglish2191346.413,375N/AAug 6, 2026
textindustry entertainment and sports and media2211259.54,566N/AAug 6, 2026
text style controloverall2211327.326,448N/AAug 6, 2026
text style controlindustry medicine and healthcare2211337.01,621N/AAug 6, 2026
text style controlexclude ties2221276.018,327N/AAug 6, 2026
text style controlnon english2221303.713,064N/AAug 6, 2026
text style controlrussian2231306.71,898N/AAug 6, 2026
text style controlmulti turn2241320.14,514N/AAug 6, 2026
text style controlinstruction following2261311.66,120N/AAug 6, 2026
text style controlindustry writing and literature and language2271304.85,958N/AAug 6, 2026
text style controlcreative writing2341284.33,361N/AAug 6, 2026
text style controlindustry life and physical and social science2341332.94,606N/AAug 6, 2026
text style controlindustry entertainment and sports and media2441271.14,566N/AAug 6, 2026
text style controlindustry legal and government2481322.91,770N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
N/A
Official provider
N/A
Lowest third-party
From $0.08 input, $0.29 output per 1M via Helicone
Tracked offerings
8
7 rows
Columns

Show columns

Heliconeqwen3-30b-a3bglobal$0.08$0.2941KAug 7, 2026
NovitaAIqwen/qwen3-30b-a3b-fp8global$0.09$0.4541KAug 7, 2026
Jiekou.AIqwen/qwen3-30b-a3b-fp8global$0.09$0.4541KAug 7, 2026
NanoGPTqwen/qwen3-30b-a3bglobal$0.10$0.3041KAug 7, 2026
302.AIqwen3-30b-a3bglobal$0.11$1.08128KAug 7, 2026
Vercel AI Gatewayalibaba/qwen-3-30bglobal$0.12$0.5041KAug 7, 2026
OpenRouterqwen/qwen3-30b-a3bglobal$0.12$0.50131.1KAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

Qwen3 30B A3B versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

Qwen3 30B A3BApr 29, 202539.730.5B128K128KNoApache 2.0

What is Qwen3 30B A3B?

A concise description based on the published model registry.

Qwen3-30B-A3B is a smaller Mixture-of-Experts (MoE) model from the Qwen3 series by Alibaba, with 30.5 billion total parameters and 3.3 billion activated parameters. Features hybrid thinking/non-thinking modes, support for 119 languages, and enhanced agent capabilities. It aims to outperform previous models like QwQ-32B while using significantly fewer activated parameters.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

Qwen3 30B A3B vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

Qwen3 30B A3BvsQwen3 MaxQwen3 30B A3BvsHermes 3 70BQwen3 30B A3BvsClaude Sonnet 3Qwen3 30B A3BvsGemma 3 12BQwen3 30B A3BvsJamba 1.5 LargeQwen3 30B A3BvsDiffusionGemma 26B A4B

Models similar to Qwen3 30B A3B

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#112+0.6
AC

Qwen3 Max

Alibaba Cloud / Qwen Team

40.3 LLMBoard

DetailsCompare
#111+0.6
AC

Qwen3 VL 30B A3B Thinking

Alibaba Cloud / Qwen Team

40.3 LLMBoard

DetailsCompare
#120-1.7
AC

Qwen2 72B

Alibaba Cloud / Qwen Team

38.0 LLMBoard

DetailsCompare
#108+2.1
AC

Qwen3 Next 80B A3B

Alibaba Cloud / Qwen Team

41.8 LLMBoard

DetailsCompare
#106+3.0
AC

Qwen3 VL 235B A22B

Alibaba Cloud / Qwen Team

42.7 LLMBoard

DetailsCompare
#124-3.6
AC

Qwen3.5 9B

Alibaba Cloud / Qwen Team

36.1 LLMBoard

DetailsCompare

FAQ

Common questions about Qwen3 30B A3B.

When was Qwen3 30B A3B released?

Qwen3 30B A3B's default version was released on Apr 29, 2025.

How much does Qwen3 30B A3B cost?

No official standard PAYG price is currently available for Qwen3 30B A3B. The lowest tracked third-party offer starts at $0.08 input and $0.29 output via Helicone.

Who created Qwen3 30B A3B?

Qwen3 30B A3B is published under Alibaba Cloud / Qwen Team in the model registry.

What is the context window for Qwen3 30B A3B?

The default version has a 128K token context window.

Is Qwen3 30B A3B open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer Qwen3 30B A3B?

8 published provider offerings are linked to the default version.

What models should I compare Qwen3 30B A3B with?

Nearby ranked alternatives include Qwen3 Max, Hermes 3 70B, Claude Sonnet 3.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai