PoolsideOpen weightsReasoning modelAuto-detected

Laguna M.1

Name: Laguna M.1 benchmark results
Creator: BenchLeader

Best configuration ranks #560 of 610 on the BenchLeader Index at 37.5 ±9.7. Last measured 8 Sept 2026. Released 28 Apr 2026.

How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Benchmark	default	Source	Trend
GPQA Diamond (Vals)not in index	27.0%#129	Vals AI

Benchmark	default	Source
LMArena WebDev	1347#91	LMArena
LiveCodeBench	68.1%#90	Vals AI
SWE-bench (Vals)not in index	57.6%#76	Vals AI

Benchmark	default	Source	Trend
Terminal-Bench 2.1 (Vals)	34.1%#58	Vals AI

Benchmark	default	Source	Trend
ProofBench	0.0%#60	Vals AI

Benchmark	default	Source
MMLU-Pro	68.8%#119	Vals AI
LegalBench	75.1%#106	Vals AI
CorpFin	58.2%#83	Vals AI
TaxEval	1.6%#137	Vals AI