Qwen 3 Coder Plus: benchmark results
Alibaba's hosted agentic-coding API tier built on Qwen3-Coder 480B A35B MoE, with up to 1M context. Provider: Alibaba. Released 2025-09-23. Access: API.
Aggregate Rating 1637 ± 11, rank #338 of 1656 rated models, from 54 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| WebCoderBench - Performance | 98.54 | Score (%) | 100 |
| BIRD-Python | 61.6 | LLM-validated execution accuracy (%; generated pandas code e | 80 |
| BenchTable - Tech | 75.4 | Weighted Score (%) | 79.7 |
| IndustryCode - Python | 31.7 | Pass@1 (%) on the Python main problems of the IndustryCode t | 77.5 |
| BacktestBench - Parameter Confirmation | 71.08 | Back-test accuracy (%) on the Parameter Confirmation task fa | 72.7 |
| BacktestBench - Strategy Selection | 77.91 | Back-test accuracy (%) on the Strategy Selection task family | 72.7 |
| BenchTable - STEM | 64.6 | Weighted Score (%) | 70.6 |
| IndustryCode - C++ | 36 | Pass@1 (%) on the C++ main problems of the IndustryCode test | 70 |
| Tinybird AI SQL Benchmark - Exactness | 51.39 | Result exactness vs human reference queries (0-100) | 67.7 |
| Tinybird AI SQL Benchmark - First-Attempt Success Rate | 96 | Questions answered with a valid query on the first attempt ( | 65.9 |
| SnakeBench | 24.9 | TrueSkill Rating | 64.2 |
| BacktestBench - SQL Execution Accuracy | 94.03 | Execution accuracy (%) of the intermediate SQL: the share of | 59.1 |
Interactive version: theaggregate.ai/model?slug=qwen-3-coder-plus · How It Works · Data refreshed daily, snapshot 2026-10-20.