Qwen 3 Next 80B A3B — benchmark results
Alibaba Qwen 3 Next 80B A3B efficient MoE model (3B active). Provider: Alibaba. Released 2025-09-11. Access: Open.
Unified ELO 1739 ± 81, rank #212 of 1776 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| XDomainBench | 63.8 | R (k=1, Deterministic) (self-reported) | 100 |
| Ko-AgentBench - L1 Single Tool Call | 77.27 | Argument Accuracy (%) | 92.9 |
| Ko-AgentBench - L3 Sequential Tool Reasoning | 100 | PSM (%) | 82.1 |
| Ko-AgentBench - L4 Parallel Tool Reasoning | 70 | Coverage (%) | 78.6 |
| DVMap | 47.6 | Accuracy (self-reported) | 75 |
| Ko-AgentBench - L5 Error Handling & Robustness | 25.42 | Adaptive Routing Score (%) | 64.3 |
| Ko-AgentBench - L7 Long-Context Memory | 97.5 | Context Retention (%) | 57.1 |
| Context Arena MRCR (2-needle) | 10.3 | AUC@1M (%) | 54.4 |
| LLM2014 Logic 2025-10 | 38.87 | Median Score | 51 |
| BacktestBench | 44.06 | Overall Accuracy (OA) (self-reported) | 50 |
| LLM2014 Logic 2025-11 | 35.65 | Median Score | 44.2 |
| LLM Chess (Saplin) | -127.9 | ELO | 35 |
Interactive version: theaggregate.ai/model?slug=qwen-3-next-80b-a3b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.