Benchmark Almanac

Laguna XS.2

Laguna XS.2.

Poolside Released April 28th, 2026 Compare with others
17
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Poolside Apr 2026

Laguna XS.2

Open weights

No Epoch Capabilities Index score yet
  • Harvey LAB (Legal Agent Benchmark) 0% #75 /80
  • Vals Finance Agent v2 15.6% #79 /80
  • Legal Research Bench 0.96% #92 /93

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Math and science

BenchmarkScoreRankSource
ProofBench 0% #59 of 59 Official leaderboard

Coding

BenchmarkScoreRankSource
SWE-bench Verified 69.9% #80 of 155 Lab model card
SWE-bench Multilingual 62.4% #58 of 73 Lab model card
Code Arena 1,303 #109 of 128 Official leaderboard
WebDev Arena 1,302 #90 of 105 arena.ai, via Epoch AI
Vibe Code Bench 5.21% #105 of 120 Official leaderboard
SWE-Bench Pro 46.3% #75 of 81 Lab model card

Agents and tools

BenchmarkScoreRankSource
Terminal-Bench 2.0 35.7% #57 of 63 Lab model card

Professional work

BenchmarkScoreRankSource
CorpFin v2 56.3% #84 of 119 Official leaderboard
LegalBench 71% #112 of 138 Official leaderboard
TaxEval v2 59% #118 of 129 Official leaderboard
Harvey LAB (Legal Agent Benchmark) 0% #75 of 80 Official leaderboard
MedScribe 61.4% #92 of 97 Official leaderboard
Vals Finance Agent v2 15.6% #79 of 80 Official leaderboard
Legal Research Bench 0.96% #92 of 93 Official leaderboard
MedCode 21.3% #95 of 96 Official leaderboard
Public Benefits Bench 41.4% #50 of 50 Official leaderboard

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.

Its system card

Every result in the longer documents the lab published with it.