Alibaba Cloud / Qwen Team model product
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
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version Qwen3 30B A3B.
| Arena Hard | 0.9 | 3 | 26 | 92.0% | C | |
| BFCL | 0.7 | 7 | 11 | 40.0% | C | |
| Multi-IF | 0.7 | 13 | 20 | 36.8% | C | |
| LiveBench | 0.7 | 18 | 38 | 54.0% | C | |
| AIME 2024 | 0.8 | 27 | 53 | 50.0% | C | |
| LiveCodeBench | 0.6 | 31 | 73 | 58.3% | C | |
| AIME 2025 | 0.7 | 87 | 114 | 23.9% | C | |
| GPQA | 0.7 | 146 | 233 | 37.5% | C |
Preference and agent-evaluation signals from published Arena datasets.
| text | japanese | 157 | 1251.9 | 665 | N/A | |
| text | french | 161 | 1353.3 | 322 | N/A | |
| text | korean | 168 | 1262.6 | 439 | N/A | |
| text style control | french | 169 | 1364.6 | 322 | N/A | |
| text | industry mathematical | 171 | 1364.5 | 1,577 | N/A | |
| text | german | 172 | 1308.3 | 727 | N/A | |
| text style control | japanese | 172 | 1251.5 | 665 | N/A | |
| text | spanish | 174 | 1318.6 | 410 | N/A | |
| text style control | industry mathematical | 175 | 1367.1 | 1,577 | N/A | |
| text | polish | 176 | 1299.9 | 1,956 | N/A | |
| text | chinese | 181 | 1355.5 | 1,252 | N/A | |
| text | math | 183 | 1355.1 | 1,701 | N/A | |
| text style control | korean | 184 | 1268.1 | 439 | N/A | |
| text style control | math | 186 | 1352.6 | 1,701 | N/A | |
| text style control | polish | 188 | 1303.9 | 1,956 | N/A | |
| text style control | spanish | 188 | 1324.5 | 410 | N/A | |
| text style control | chinese | 189 | 1370.0 | 1,252 | N/A | |
| text style control | german | 190 | 1304.1 | 727 | N/A | |
| text | expert | 193 | 1313.9 | 1,313 | N/A | |
| text style control | expert | 195 | 1343.0 | 1,313 | N/A | |
| text style control | coding | 199 | 1386.5 | 4,526 | N/A | |
| text | coding | 202 | 1338.1 | 4,526 | N/A | |
| text | industry business and management and financial operations | 202 | 1308.8 | 4,010 | N/A | |
| text | industry medicine and healthcare | 202 | 1317.9 | 1,621 | N/A | |
| text | industry software and it services | 202 | 1340.5 | 8,301 | N/A | |
| text | longer query | 205 | 1313.3 | 4,446 | N/A | |
| text | non english | 205 | 1294.8 | 13,064 | N/A | |
| text | hard prompts | 206 | 1315.5 | 10,682 | N/A | |
| text | hard prompts english | 207 | 1331.4 | 5,806 | N/A | |
| text | industry legal and government | 207 | 1313.7 | 1,770 | N/A | |
| text | industry life and physical and social science | 207 | 1321.1 | 4,606 | N/A | |
| text style control | industry business and management and financial operations | 207 | 1330.0 | 4,010 | N/A | |
| text style control | industry software and it services | 207 | 1371.4 | 8,301 | N/A | |
| text | english | 208 | 1335.3 | 13,375 | N/A | |
| text | russian | 208 | 1288.8 | 1,898 | N/A | |
| text style control | hard prompts english | 208 | 1362.3 | 5,806 | N/A | |
| text style control | longer query | 208 | 1339.5 | 4,446 | N/A | |
| text | overall | 209 | 1317.2 | 26,448 | N/A | |
| text | exclude ties | 210 | 1260.8 | 18,327 | N/A | |
| text | multi turn | 210 | 1307.5 | 4,514 | N/A | |
| text style control | hard prompts | 211 | 1345.4 | 10,682 | N/A | |
| text | industry writing and literature and language | 213 | 1291.1 | 5,958 | N/A | |
| text | instruction following | 214 | 1285.7 | 6,120 | N/A | |
| text | creative writing | 219 | 1272.2 | 3,361 | N/A | |
| text style control | english | 219 | 1346.4 | 13,375 | N/A | |
| text | industry entertainment and sports and media | 221 | 1259.5 | 4,566 | N/A | |
| text style control | overall | 221 | 1327.3 | 26,448 | N/A | |
| text style control | industry medicine and healthcare | 221 | 1337.0 | 1,621 | N/A | |
| text style control | exclude ties | 222 | 1276.0 | 18,327 | N/A | |
| text style control | non english | 222 | 1303.7 | 13,064 | N/A | |
| text style control | russian | 223 | 1306.7 | 1,898 | N/A | |
| text style control | multi turn | 224 | 1320.1 | 4,514 | N/A | |
| text style control | instruction following | 226 | 1311.6 | 6,120 | N/A | |
| text style control | industry writing and literature and language | 227 | 1304.8 | 5,958 | N/A | |
| text style control | creative writing | 234 | 1284.3 | 3,361 | N/A | |
| text style control | industry life and physical and social science | 234 | 1332.9 | 4,606 | N/A | |
| text style control | industry entertainment and sports and media | 244 | 1271.1 | 4,566 | N/A | |
| text style control | industry legal and government | 248 | 1322.9 | 1,770 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| Helicone | qwen3-30b-a3b | global | $0.08 | $0.29 | 41K | |
| NovitaAI | qwen/qwen3-30b-a3b-fp8 | global | $0.09 | $0.45 | 41K | |
| Jiekou.AI | qwen/qwen3-30b-a3b-fp8 | global | $0.09 | $0.45 | 41K | |
| NanoGPT | qwen/qwen3-30b-a3b | global | $0.10 | $0.30 | 41K | |
| 302.AI | qwen3-30b-a3b | global | $0.11 | $1.08 | 128K | |
| Vercel AI Gateway | alibaba/qwen-3-30b | global | $0.12 | $0.50 | 41K | |
| OpenRouter | qwen/qwen3-30b-a3b | global | $0.12 | $0.50 | 131.1K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| Qwen3 30B A3B | 39.7 | 30.5B | 128K | 128K | No | Apache 2.0 |
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.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Qwen3 30B A3B.
Qwen3 30B A3B's default version was released on Apr 29, 2025.
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.
Qwen3 30B A3B is published under Alibaba Cloud / Qwen Team in the model registry.
The default version has a 128K token context window.
No. The default version is not marked as having publicly available weights.
8 published provider offerings are linked to the default version.
Nearby ranked alternatives include Qwen3 Max, Hermes 3 70B, Claude Sonnet 3.