Laguna XS.2 · ProofBench · Accuracy · % ↗
Poolside / MODEL EVIDENCE
Laguna XS.2
17 benchmarks · 17 observations · 17 sources · released 2026-04-28
Scores at a glance
17 benchmarks · large dot: Laguna XS.2; small dots: closest models where reported
Text Arena (Coding) ↗1302.4 Elo
Laguna XS.2 · WebDev Arena · Elo ↗
Legal Research Bench ↗0.96 %
Laguna XS.2 · Legal Research Bench · All-Pass ↗
LegalBench ↗71.03 %
Laguna XS.2 · LegalBench · Accuracy ↗
Laguna XS.2 · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Vals Finance Agent v2 ↗15.6 %
Laguna XS.2 · Finance Agent v2 · Accuracy ↗
CorpFin v2 ↗56.33 %
Laguna XS.2 · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗58.95 %
Laguna XS.2 · TaxEval · Accuracy ↗
MedCode ↗21.25 %
Laguna XS.2 · MedCode · Accuracy ↗
MedScribe ↗61.43 %
Laguna XS.2 · MedScribe · Accuracy ↗
GPQA Diamond ↗55.05 %
Laguna XS.2 · GPQA Diamond · Accuracy ↗
MMLU-Pro ↗69.05 %
Laguna XS.2 · MMLU Pro · Accuracy ↗
Vibe Code Bench ↗5.21 %
Laguna XS.2 (OpenHands) · Vibe Code Bench · original task set ↗
Laguna XS.2 · ProgramBench · Accuracy ↗
SWE-bench Verified ↗55.2 %
Laguna XS.2 (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗67.8 %
Laguna XS.2 · LiveCodeBench · Accuracy ↗
Public Benefits Bench v1.1 ↗41.41 %
Laguna XS.2 · Public Benefits Bench · Accuracy ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
13 of 15 shared benchmark comparisons
Mistral Medium 3.5 →Mistral
9 of 15 shared benchmark comparisons
Claude Haiku 4.5 →Anthropic
13 of 15 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Arena leaderboard — WebDev evaluation ↗
- CorpFin — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedCode — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- Public Benefits Bench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Names used in source tables
Laguna XS.2 · Laguna XS.2 (Mini-SWE-agent) · Laguna XS.2 (OpenHands)