Atlas/Models/claude-opus-4-5

Exact model identity

Claude Opus 4.5

Anthropic · version status: reported

Similar model names are never merged automatically. Evaluation membership and numeric result rows only reference the exact ID claude-opus-4-5.

Evaluation settings

Benchmark	Work	Run / group	Scope
Anthropic Computational Biology Eval	Advancing Claude in healthcare and the life sciences	anthropic-computational-biology-delta anthropic-computational-biology-delta	unknown
Anthropic Protein Understanding Eval	Advancing Claude in healthcare and the life sciences	anthropic-protein-understanding-delta anthropic-protein-understanding-delta	unknown
Anthropic Scientific Figure Interpretation Eval	Advancing Claude in healthcare and the life sciences	anthropic-scientific-figure-delta anthropic-scientific-figure-delta	unknown
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-base spatialbench-paper-v2-base	full · n=146
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-claude-code spatialbench-paper-v2-claude-code	full · n=146
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-latch spatialbench-paper-v2-latch	full · n=146
SpatialBench	SpatialBench 159-evaluation repository snapshot	spatialbench-repo-159-mini-swe-agent spatialbench-repo-159-mini-swe-agent	full · n=159

Numeric results

Benchmark	Work	Run / group	Metric	Value
Anthropic Computational Biology Eval	Advancing Claude in healthcare and the life sciences	anthropic-computational-biology-delta anthropic-computational-biology-delta	Accuracy improvement from Claude Opus 4.1 to Claude Opus 4.5 delta vs claude-opus-4-1	Δ 10.5 percent delta as annotated
Anthropic Protein Understanding Eval	Advancing Claude in healthcare and the life sciences	anthropic-protein-understanding-delta anthropic-protein-understanding-delta	Accuracy improvement from Claude Opus 4.1 to Claude Opus 4.5 delta vs claude-opus-4-1	Δ 10.3 percent delta as annotated
Anthropic Scientific Figure Interpretation Eval	Advancing Claude in healthcare and the life sciences	anthropic-scientific-figure-delta anthropic-scientific-figure-delta	Accuracy improvement from Claude Opus 4.1 to Claude Opus 4.5 delta vs claude-opus-4-1	Δ 13.2 percent delta as annotated
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-base spatialbench-paper-v2-base	Accuracy	38.36 percent
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-base spatialbench-paper-v2-base	Steps	2.84 steps per evaluation
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-base spatialbench-paper-v2-base	Latency	123.8 seconds per evaluation
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-base spatialbench-paper-v2-base	Cost	0.143 USD per evaluation
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-claude-code spatialbench-paper-v2-claude-code	Accuracy	48.1 percent
SpatialBench	SpatialBench: Can Agents Analyze Real-World Spatial Biology Data?	spatialbench-paper-v2-latch spatialbench-paper-v2-latch	Accuracy	61.7 percent
SpatialBench	SpatialBench 159-evaluation repository snapshot	spatialbench-repo-159-mini-swe-agent spatialbench-repo-159-mini-swe-agent	Accuracy	42.77 percent
SpatialBench	SpatialBench 159-evaluation repository snapshot	spatialbench-repo-159-mini-swe-agent spatialbench-repo-159-mini-swe-agent	Cost	0.4624 USD per evaluation
SpatialBench	SpatialBench 159-evaluation repository snapshot	spatialbench-repo-159-mini-swe-agent spatialbench-repo-159-mini-swe-agent	Duration	376.54 seconds per evaluation