Frontier Log

Benchmark · Active

SWE-bench Pro

Created by Scale AI, 19 Sep 2025 · Higher is better · Website

0102030405060Apr 2026Jul 202649.2%Laguna M.147.6%Laguna XS 2.144.5%Laguna XS.20102030405060Apr 2026Jul 202649.2%Laguna M.147.6%Laguna XS 2.144.5%Laguna XS.2

Settings and run setups

Task setProviders
Public set (731 tasks)OpenAIAnthropicGoogleMetaMoonshot AINVIDIACognitionThinking Machines
Alibaba's corrected setAlibaba
Meituan's corrected setMeituan
Poolside's patched public setPoolside
Not statedOpenAIAnthropicDeepSeekAlibabaMoonshot AIMiniMaxZ.aiNVIDIATencentCognitionRicursive IntelligencePoolside

Download image

Results

ModelScoreSource
Laguna M.128 Apr 202649.2%Long context update: Laguna XS.2 and M.1Corroborated The same number is printed in 1 other lab document: 1poolside harness on Harbor · pass@1 averaged over 3 runs · some task images patched · public set per the model card · reported laterOur capture
Laguna XS 2.12 Jul 202647.6%Introducing Laguna XS 2.1Checked A person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.poolside harness on Harbor · pass@1 averaged over 2 attempts · some task images patched · public set per the model cardOur capture
Laguna XS.228 Apr 202644.5%Introducing Laguna XS.2 and Laguna M.1Corroborated The same number is printed in 1 other lab document: 1poolside harness on Harbor · pass@1 averaged over 3 runs · some task images patched · public set per the model cardOur capture
Laguna XS.228 Apr 202646.3%Long context update: Laguna XS.2 and M.1Corroborated The same number is printed in 2 other lab documents: 1 2poolside harness on Harbor · pass@1 averaged over 3 runs · some task images patched · public set per the model card · reported later · another report, not chartedOur capture

All benchmarks