Skip to main content
Benchmark Registry
ModelsBenchmarksOrganizations

DeepSWE

Benchmark
DeepSWE
Evaluated by
DataCurve
Release date
June 15, 2026
Version
DeepSWE 1.1
Metric
DeepSWE Pass@1

Results

5 results

LatestHistory
All providersAlibaba (Tongyi)AnthropicDeepSeek AIGoogle (DeepMind)Moonshot AIOpenAISpaceXAIZ.ai
DeepSWE 1.1 results
ProviderModelScoreSourceRegistry No.
AnthropicClaude Sonnet 5.5 (max)71.0%Source for Claude Sonnet 5.5 (max) on DeepSWE 1.1 (opens in a new tab)20016
AnthropicClaude Opus 4.8 (max)59.0%Source for Claude Opus 4.8 (max) on DeepSWE 1.1 (opens in a new tab)20011
AnthropicClaude Fable 5 (xhigh)70.0%Source for Claude Fable 5 (xhigh) on DeepSWE 1.1 (opens in a new tab)20012
AnthropicClaude Sonnet 5 (max)54.0%Source for Claude Sonnet 5 (max) on DeepSWE 1.1 (opens in a new tab)20013
AnthropicClaude Opus 5 (max)74.0%Source for Claude Opus 5 (max) on DeepSWE 1.1 (opens in a new tab)20014
Page 1 of 1

© 2026 Densa Labs

Toggle between the data update date and the application build time.
Legal
Color theme