Atlas

Models

← All models

Qwen3.5-27B

Alibaba · 2026-02-24 · 27.8B parameters

Qwen3.5-27B is Alibaba's open-weight, dense, post-trained multimodal model in the Qwen3.5 medium series, released February 24, 2026. Its complete checkpoint contains 27,781,427,952 parameters; it accepts text, images, and video, produces text, uses a hybrid Gated DeltaNet/full-attention architecture, supports optional thinking, and is Apache 2.0 licensed. It supports 262,144 tokens natively and documents YaRN extension to 1,010,000 tokens.

Benchmark scores

BenchmarkScore
AA-LCR67.3
AA-Omniscience Index-42.0
AIME 202691.7
ALE-Bench882
Apex (MathArena)2.1
Apex Shortlist52.1
Artificial Analysis Intelligence Index33.8
Artificial Analysis Omniscience Accuracy21.0
Artificial Analysis Omniscience Hallucination Rate74.1
Artificial Analysis Openness Index38.9
ArXivMath 01/202653.3
ArXivMath 02/202631.3
ArXivMath 12/202541.2
BuseyBench SVG2.8
Capability146.8
CritPt0.9
Eval Connections60.0
Eval Connections — Guess Accuracy57.7
Extended NYT Connections 940-Puzzle Set60.8
Extended NYT Connections Newest 10058.2
Final-Answer Comps — Overall56.7
GBENCH One-shot Coding GScore30.2
Global MMLU Lite86.8
Global PIQA - Non-Parallel (Strict Exact Match)78.5
Global PIQA - Parallel (Strict Exact Match)71.6
GPQA Diamond85.8
HMMT Feb 202681.1
Humanity's Last Exam (Text-Only)22.2
IFBench (Artificial Analysis)75.6
ITBench-AA35.5
Loading Atlas data…