ITBench-AA ↗35.50%
Qwen3.5 27B (Reasoning) · ITBench-AA · root-cause score ↗
Qwen / MODEL EVIDENCE
Qwen3.5 27B (Reasoning)
3 benchmarks · 3 observations · 3 sources · released 2026-02-24
Scores at a glance
3 benchmarks · large dot: Qwen3.5 27B (Reasoning); small dots: closest models where reported
IFBench ↗75.58%
Qwen3.5 27B (Reasoning) · IFBench · AA implementation ↗
Global-MMLU-Lite ↗86.76%
Qwen3.5 27B (Reasoning) · Global-MMLU-Lite ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
There are not enough shared scored settings to identify close comparisons reliably.
POST-TRAINING / SOURCE DISCLOSURE
What the Qwen3.5 post-trained family report says
The reviewed card describes progressively harder agent environments and asynchronous reinforcement-learning infrastructure. Its architecture and multimodal training are documented separately from its post-training objective.
- Agent environments with increasing task complexity
- Large-scale asynchronous reinforcement learning
The card reviewed is for 397B-A17B. It does not name a complete policy objective; the background links do not establish that every Qwen3.5 checkpoint used the same optimizer.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Global-MMLU-Lite Benchmark | Artificial Analysis ↗
- IFBench Benchmark | Artificial Analysis ↗
- ITBench-AA Benchmark | Artificial Analysis ↗
Names used in source tables
Qwen3.5 27B (Reasoning)