Atlas

Models

← All models

Claude Sonnet 5.5

Anthropic · 2026-09-28

Claude Sonnet 5.5 is Anthropic's September 28, 2026 model for coding and professional work. Adaptive thinking is on by default with high effort; low, medium, high, xhigh, and max are supported. The between_tools setting disables up-front thinking at high effort or below. Provider fallbacks can route declined requests to Sonnet 5. Its published June 2026 knowledge cutoff has no day precision.

Benchmark scores

BenchmarkScore
AA-Briefcase v1.1 Elo1811
AA-LCR v1.182.7
AA-Omniscience Index32.3
ActiveVision41.2
ActiveVision with tools78.8
APEX (Mercor)17.5
APEX-Agents 1.183.7
APEX-Agents 1.1 Pass@175.5
APEX-SWE66.4
APEX-v1-extended53.9
Artificial Analysis Intelligence Index v4.3.256.0
Artificial Analysis Omniscience Accuracy54.0
Artificial Analysis Omniscience Hallucination Rate47.0
AutomationBench-AA 1.0.671.3
BioMysteryBench — Hard (Vals)45.1
BioMysteryBench — Human-solvable (Vals)89.5
BioMysteryBench (Vals Terminus 2)81.1
BuseyBench SVG6.5
Capability165.5
Code Migration69.8
Code Migration - CLI69.8
Code Migration - COBOL70.0
Code Migration — Code Quality4.6
Code Migration (Vals Index v2 subset)70.1
CritPt (avg@5)31.4
CursorBench 4.055.5
Excel Modeling Benchmark75.7
Excel Modeling Benchmark Comps80.0
Excel Modeling Benchmark Dataroom86.3
Excel Modeling Benchmark DCF70.1
Loading Atlas data…