Alibaba Cloud / Qwen Team model product
Qwen3.5-122B-A10B is a multimodal Mixture-of-Experts model with 122 billion total parameters and 10 billion activated parameters. It combines strong reasoning, coding, long-context, and visual understanding performance with production-friendly efficiency and a native 262K context window.
Updated Aug 10, 2026. Default version: Qwen3.5-122B-A10B
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 Qwen3.5-122B-A10B.
| FullStackBench en | 0.6 | 1 | 3 | 100.0% | C | |
| FullStackBench zh | 0.6 | 1 | 3 | 100.0% | C | |
| MMBench-V1.1 | 0.9 | 1 | 18 | 100.0% | C | |
| MVBench | 0.8 | 1 | 17 | 100.0% | C | |
| Nuscene | 0.2 | 1 | 3 | 100.0% | C | |
| PMC-VQA | 0.6 | 1 | 3 | 100.0% | C | |
| SlakeVQA | 0.8 | 1 | 4 | 100.0% | C | |
| SUNRGBD | 0.4 | 1 | 4 | 100.0% | C | |
| VideoMME w/o sub. | 0.8 | 1 | 10 | 100.0% | C | |
| ZEROBench-Sub | 0.4 | 1 | 5 | 100.0% | C | |
| AndroidWorld_SR | 0.7 | 2 | 8 | 85.7% | C | |
| BrowseComp-zh | 0.7 | 2 | 13 | 91.7% | C | |
| MedXpertQA | 0.7 | 2 | 12 | 90.9% | C | |
| MLVU | 0.9 | 2 | 10 | 88.9% | C | |
| MMStar | 0.8 | 2 | 22 | 95.2% | C | |
| MMVU | 0.7 | 2 | 4 | 66.7% | C | |
| OCRBench | 0.9 | 2 | 22 | 95.2% | C | |
| DynaMath | 0.9 | 3 | 7 | 66.7% | C | |
| Hypersim | 0.1 | 3 | 4 | 33.3% | C | |
| LingoQA | 0.8 | 3 | 4 | 33.3% | C | |
| MathVista-Mini | 0.9 | 3 | 23 | 90.9% | C | |
| RefSpatialBench | 0.7 | 3 | 6 | 60.0% | C | |
| VideoMME w sub. | 0.9 | 3 | 10 | 77.8% | C | |
| BFCL-V4 | 0.7 | 4 | 13 | 75.0% | C | |
| CC-OCR | 0.8 | 4 | 18 | 82.3% | C | |
| C-Eval | 0.9 | 4 | 18 | 82.3% | C | |
| CodeForces | 0.9 | 4 | 16 | 80.0% | C | |
| Hallusion Bench | 0.7 | 4 | 16 | 80.0% | C | |
| MMLongBench-Doc | 0.6 | 4 | 5 | 25.0% | C | |
| NOVA-63 | 0.6 | 4 | 11 | 70.0% | C | |
| OJBench | 0.4 | 4 | 9 | 62.5% | C | |
| TIR-Bench | 0.5 | 4 | 4 | 0.0% | C | |
| VLMsAreBlind | 1.0 | 4 | 4 | 0.0% | C | |
| AI2D | 0.9 | 5 | 32 | 87.1% | C | |
| CountBench | 1.0 | 5 | 6 | 20.0% | C | |
| DeepPlanning | 0.2 | 5 | 9 | 50.0% | C | |
| EmbSpatialBench | 0.8 | 5 | 8 | 42.9% | C | |
| HMMT25 | 0.9 | 5 | 25 | 83.3% | C | |
| OmniDocBench 1.5 | 0.9 | 5 | 16 | 73.3% | C | |
| RefCOCO-avg | 0.9 | 5 | 7 | 33.3% | C | |
| Seal-0 | 0.4 | 5 | 6 | 20.0% | C | |
| V* | 0.9 | 5 | 7 | 33.3% | C | |
| AA-LCR | 0.7 | 6 | 15 | 64.3% | C | |
| IFEval | 0.9 | 6 | 65 | 92.2% | C | |
| Include | 0.8 | 6 | 31 | 83.3% | C | |
| LVBench | 0.7 | 6 | 24 | 78.3% | C | |
| MAXIFE | 0.9 | 6 | 11 | 50.0% | C | |
| MMLU-ProX | 0.8 | 6 | 32 | 83.9% | C | |
| MMLU-Redux | 0.9 | 6 | 48 | 89.4% | C | |
| MMMU | 0.8 | 6 | 63 | 91.9% | C | |
| PolyMATH | 0.7 | 6 | 23 | 77.3% | C | |
| Global PIQA | 0.9 | 7 | 13 | 50.0% | C | |
| MathVision | 0.9 | 7 | 32 | 80.7% | C | |
| RealWorldQA | 0.9 | 7 | 26 | 76.0% | C | |
| ScreenSpot Pro | 0.7 | 7 | 24 | 73.9% | C | |
| SuperGPQA | 0.7 | 7 | 34 | 81.8% | C | |
| VITA-Bench | 0.3 | 7 | 10 | 33.3% | C | |
| WideSearch | 0.6 | 7 | 9 | 25.0% | C | |
| ZEROBench | 0.1 | 7 | 9 | 25.0% | C | |
| BabyVision | 0.4 | 8 | 9 | 12.5% | C | |
| Multi-Challenge | 0.6 | 8 | 29 | 75.0% | C | |
| SimpleVQA | 0.6 | 8 | 13 | 41.7% | C | |
| IFBench | 0.8 | 9 | 28 | 70.4% | C | |
| ODinW | 0.4 | 9 | 16 | 46.7% | C | |
| WMT24++ | 0.8 | 9 | 23 | 63.6% | C | |
| MMLU-Pro | 0.9 | 10 | 129 | 93.0% | C | |
| ERQA | 0.6 | 11 | 23 | 54.5% | C | |
| LongBench v2 | 0.6 | 11 | 17 | 37.5% | C | |
| t2-bench | 0.8 | 13 | 23 | 45.5% | C | |
| HMMT 2025 | 0.9 | 15 | 33 | 56.3% | C | |
| VideoMMMU | 0.8 | 15 | 26 | 44.0% | C | |
| OSWorld-Verified | 0.6 | 19 | 22 | 14.3% | C | |
| LiveCodeBench v6 | 0.8 | 22 | 53 | 59.6% | C | |
| MMMLU | 0.9 | 22 | 49 | 56.3% | C | |
| Humanity's Last Exam | 0.5 | 24 | 92 | 74.7% | C | |
| MMMU-Pro | 0.8 | 25 | 65 | 62.5% | C | |
| CharXiv-R | 0.8 | 28 | 47 | 41.3% | C | |
| BrowseComp | 0.6 | 32 | 58 | 45.6% | C | |
| Terminal-Bench 2.0 | 0.5 | 36 | 49 | 27.1% | C | |
| GPQA | 0.9 | 44 | 233 | 81.5% | C | |
| SWE-Bench Verified | 0.7 | 52 | 104 | 50.5% | C |
Preference and agent-evaluation signals from published Arena datasets.
| vision | humor | 38 | 1250.8 | 411 | N/A | |
| vision | chinese | 39 | 1301.7 | 744 | N/A | |
| vision style control | chinese | 42 | 1278.7 | 744 | N/A | |
| vision style control | humor | 43 | 1227.1 | 411 | N/A | |
| vision | english | 51 | 1246.8 | 5,270 | N/A | |
| vision | homework | 51 | 1260.2 | 1,868 | N/A | |
| vision | ocr | 51 | 1251.4 | 8,693 | N/A | |
| vision | overall | 53 | 1245.6 | 12,437 | N/A | |
| vision | creative writing vision | 53 | 1228.4 | 742 | N/A | |
| vision style control | homework | 54 | 1252.5 | 1,868 | N/A | |
| vision style control | overall | 55 | 1226.9 | 12,437 | N/A | |
| vision style control | english | 55 | 1227.0 | 5,270 | N/A | |
| vision style control | ocr | 55 | 1239.3 | 8,693 | N/A | |
| vision | diagram | 56 | 1251.4 | 3,223 | N/A | |
| text factuality | chinese | 58 | 1476.8 | 1,102 | N/A | |
| vision style control | creative writing vision | 58 | 1207.4 | 742 | N/A | |
| vision style control | diagram | 59 | 1245.5 | 3,223 | N/A | |
| text factuality | industry mathematical | 62 | 1431.1 | 1,085 | N/A | |
| text factuality | math | 62 | 1430.4 | 1,208 | N/A | |
| text factuality | industry medicine and healthcare | 72 | 1462.0 | 1,373 | N/A | |
| webdev | webdev-react | 72 | 1351.8 | 6,694 | N/A | |
| webdev | overall | 76 | 1359.4 | 7,693 | N/A | |
| webdev | webdev | 76 | 1359.4 | 7,693 | N/A | |
| text factuality | expert | 77 | 1451.5 | 1,699 | N/A | |
| webdev | webdev-html | 79 | 1356.3 | 912 | N/A | |
| text | french | 80 | 1442.7 | 860 | N/A | |
| text | german | 80 | 1420.6 | 423 | N/A | |
| text factuality | industry business and management and financial operations | 81 | 1433.2 | 3,868 | N/A | |
| text factuality | industry legal and government | 81 | 1438.4 | 1,525 | N/A | |
| text | math | 84 | 1427.9 | 1,763 | N/A | |
| text factuality | russian | 85 | 1417.0 | 1,961 | N/A | |
| text | chinese | 86 | 1464.1 | 1,620 | N/A | |
| text | expert | 86 | 1435.1 | 2,461 | N/A | |
| text style control | french | 86 | 1449.2 | 860 | N/A | |
| text | spanish | 88 | 1418.8 | 849 | N/A | |
| text | industry mathematical | 89 | 1429.9 | 1,559 | N/A | |
| text | industry business and management and financial operations | 92 | 1415.2 | 5,416 | N/A | |
| text | japanese | 92 | 1365.9 | 226 | N/A | |
| text factuality | industry life and physical and social science | 93 | 1447.1 | 3,215 | N/A | |
| text factuality | coding | 94 | 1472.4 | 5,991 | N/A | |
| text | industry life and physical and social science | 96 | 1434.0 | 4,594 | N/A | |
| text factuality | instruction following | 97 | 1414.4 | 7,252 | N/A | |
| text factuality | non english | 97 | 1413.4 | 14,773 | N/A | |
| text style control | industry business and management and financial operations | 97 | 1422.5 | 5,416 | N/A | |
| text style control | german | 98 | 1413.6 | 423 | N/A | |
| text factuality | overall | 99 | 1428.3 | 28,018 | N/A | |
| text factuality | exclude ties | 99 | 1422.2 | 20,494 | N/A | |
| text factuality | industry software and it services | 99 | 1461.2 | 10,129 | N/A | |
| text | industry legal and government | 100 | 1424.8 | 2,198 | N/A | |
| text | industry medicine and healthcare | 100 | 1427.7 | 2,044 | N/A | |
| text factuality | hard prompts english | 100 | 1451.2 | 7,077 | N/A | |
| text factuality | industry writing and literature and language | 100 | 1398.1 | 4,644 | N/A | |
| text style control | math | 100 | 1422.8 | 1,763 | N/A | |
| text | coding | 101 | 1436.4 | 7,740 | N/A | |
| text factuality | english | 101 | 1437.8 | 13,108 | N/A | |
| text style control | spanish | 101 | 1411.3 | 849 | N/A | |
| text | english | 102 | 1428.7 | 13,289 | N/A | |
| text | industry software and it services | 102 | 1437.6 | 11,078 | N/A | |
| text factuality | multi turn | 102 | 1422.2 | 3,486 | N/A | |
| text style control | chinese | 102 | 1461.7 | 1,620 | N/A | |
| text style control | expert | 102 | 1445.0 | 2,461 | N/A | |
| text style control | industry mathematical | 102 | 1428.1 | 1,559 | N/A | |
| text style control | japanese | 102 | 1361.4 | 226 | N/A | |
| text | overall | 103 | 1417.8 | 28,208 | N/A | |
| text factuality | hard prompts | 103 | 1445.6 | 17,539 | N/A | |
| text | hard prompts english | 104 | 1431.0 | 8,746 | N/A | |
| text factuality | creative writing | 104 | 1380.2 | 2,997 | N/A | |
| text factuality | industry entertainment and sports and media | 104 | 1384.7 | 3,955 | N/A | |
| text factuality | longer query | 104 | 1429.2 | 9,029 | N/A | |
| text | instruction following | 105 | 1400.5 | 8,933 | N/A | |
| text | longer query | 105 | 1411.0 | 10,804 | N/A | |
| text | exclude ties | 106 | 1403.8 | 20,638 | N/A | |
| text | multi turn | 106 | 1415.1 | 4,911 | N/A | |
| text | polish | 106 | 1397.2 | 579 | N/A | |
| text | korean | 107 | 1358.2 | 447 | N/A | |
| text style control | industry life and physical and social science | 107 | 1437.1 | 4,594 | N/A | |
| text | hard prompts | 109 | 1421.8 | 17,671 | N/A | |
| text | non english | 110 | 1401.1 | 14,919 | N/A | |
| text style control | english | 110 | 1430.4 | 13,289 | N/A | |
| text style control | instruction following | 110 | 1407.1 | 8,933 | N/A | |
| text style control | industry legal and government | 112 | 1425.7 | 2,198 | N/A | |
| text | industry writing and literature and language | 113 | 1388.4 | 6,385 | N/A | |
| text style control | overall | 113 | 1416.9 | 28,208 | N/A | |
| text style control | exclude ties | 113 | 1404.3 | 20,638 | N/A | |
| text style control | hard prompts english | 113 | 1443.5 | 8,746 | N/A | |
| text style control | korean | 113 | 1349.9 | 447 | N/A | |
| text style control | multi turn | 115 | 1418.5 | 4,911 | N/A | |
| text style control | coding | 117 | 1459.2 | 7,740 | N/A | |
| text style control | industry software and it services | 117 | 1450.7 | 11,078 | N/A | |
| text | russian | 118 | 1396.7 | 2,955 | N/A | |
| text style control | hard prompts | 119 | 1432.8 | 17,671 | N/A | |
| text style control | non english | 120 | 1397.8 | 14,919 | N/A | |
| text | creative writing | 122 | 1369.2 | 4,272 | N/A | |
| text style control | industry writing and literature and language | 122 | 1388.8 | 6,385 | N/A | |
| text style control | longer query | 122 | 1418.5 | 10,804 | N/A | |
| text | industry entertainment and sports and media | 123 | 1368.0 | 5,409 | N/A | |
| text style control | industry medicine and healthcare | 125 | 1431.7 | 2,044 | N/A | |
| text style control | polish | 126 | 1383.7 | 579 | N/A | |
| text style control | russian | 132 | 1395.6 | 2,955 | N/A | |
| text style control | creative writing | 138 | 1367.0 | 4,272 | 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.
| NanoGPT | qwen3.5-122b-a10b | global | $0.437 | $3.5 | 131.1K | |
| Cortecs | qwen3.5-122b-a10b | global | $0.495 | $3.46 | 262.1K |
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.
| Qwen3.5-122B-A10B | 61.4 | 122B | 262.1K | 65.5K | Yes | Apache 2.0 |
A concise description based on the published model registry.
Qwen3.5-122B-A10B is a multimodal Mixture-of-Experts model with 122 billion total parameters and 10 billion activated parameters. It combines strong reasoning, coding, long-context, and visual understanding performance with production-friendly efficiency and a native 262K context window.
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 Qwen3.5 122B A10B.
Qwen3.5 122B A10B's default version was released on Feb 24, 2026.
No official standard PAYG price is currently available for Qwen3.5 122B A10B. The lowest tracked third-party offer starts at $0.437 input and $3.5 output via NanoGPT.
Qwen3.5 122B A10B is published under Alibaba Cloud / Qwen Team in the model registry.
The default version has a 262.1K token context window.
Yes. The default version is marked as open weight under Apache 2.0.
2 published provider offerings are linked to the default version.
Nearby ranked alternatives include Llama 3.1 Nemotron Ultra 253B, GLM 4.5, o3 mini.