Qwen3-Next-80B-A3B-Instruct
Zero-eval
#1Arena-Hard v2
#2MultiPL-E
#2WritingBench
+3 more
by Alibaba Cloud / Qwen Team
About
Qwen3-Next-80B-A3B-Instruct is a language model developed by Alibaba Cloud / Qwen Team. It achieves strong performance with an average score of 67.0% across 24 benchmarks. It excels particularly in MMLU-Redux (90.9%), MultiPL-E (87.8%), IFEval (87.6%). It supports a 131K token context window for handling large documents. The model is available through 1 API provider. It's licensed for commercial use, making it suitable for enterprise applications. Released in 2025, it represents Alibaba Cloud / Qwen Team's latest advancement in AI technology.
Pricing Range
Input (per 1M)$0.15 -$0.15
Output (per 1M)$1.50 -$1.50
Providers1
Timeline
AnnouncedJan 10, 2025
ReleasedJan 10, 2025
Specifications
Training Tokens15.0T
License & Family
License
Apache 2.0
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
24 benchmarks
Average Score
67.0%
Best Score
90.9%
High Performers (80%+)
7Performance Metrics
Max Context Window
131.1KTop Categories
factuality
90.9%
general
75.8%
roleplay
75.8%
code
70.4%
reasoning
58.8%
Additional Information
Content coming soon...
All Benchmark Results for Qwen3-Next-80B-A3B-Instruct
Complete list of benchmark scores with detailed information
| MMLU-Redux | factuality | text | 0.91 | 90.9% | Self-reported |
| MultiPL-E | code | text | 0.88 | 87.8% | Self-reported |
| IFEval | code | text | 0.88 | 87.6% | Self-reported |
| WritingBench | general | text | 0.87 | 87.3% | Self-reported |
| Creative Writing v3 | general | text | 0.85 | 85.3% | Self-reported |
| Arena-Hard v2 | general | text | 0.83 | 82.7% | Self-reported |
| MMLU-Pro | general | text | 0.81 | 80.6% | Self-reported |
| Include | general | text | 0.79 | 78.9% | Self-reported |
| MMLU-ProX | general | text | 0.77 | 76.7% | Self-reported |
| LiveBench 20241125 | roleplay | text | 0.76 | 75.8% | Self-reported |
Showing 1 to 10 of 24 benchmarks