Which model leads SWE-bench Multimodal?

Q: Which model leads SWE-bench Multimodal?

Claude Opus 5.5 tops the board at 61.4

SWE-bench Multimodal Leaderboard — Model Scores

Pro Mode — raw scores, one benchmark under the lensofficial card harness

The field · 14 measured

19.861.4

14 models measured, most on the official card harness. Claude Opus 5.5 tops the board at 61.4.

14 measured·2 new this week·Lifecycle Updated Oct 8, 2026Score basis

#◆Vendor$/M in$/M out

Leader — ignites whiteField — fill cools with scoreIndependently verified only0 of 14Bars show 19.8–61.4 · ◆ solid = first-party or better · outlined = arm's-length

01Claude Opus 5.5+2 altsAnthropic61.4$4.0/M$20.0/M$0.195

02Claude Opus 5+2 altsAnthropic59.4$5.0/M$25.0/M$0.253

03Claude Mythos PreviewAnthropic59———

04Claude Mythos 5.1Anthropic54.7$10.0/M$50.0/M$0.548

05Claude Fable 5.1+1 altAnthropic54.7$10.0/M$50.0/M$0.548

06Sonnet 5.5+2 altsAnthropic54.3$2.0/M$10.0/M$0.110

07Claude Fable 5+1 altAnthropic54.1$10.0/M$50.0/M$0.555

08Qwen3.8 27BAlibaba38.6$0.50/M$3.0/M$0.045

09Claude Opus 4.8+2 altsAnthropic38.4$5.0/M$25.0/M$0.391

10Claude Opus 4.7+2 altsAnthropic34.5$5.0/M$25.0/M$0.435

11Haiku 5.5NEW+1 altAnthropic30.7$0.10/M$0.50/M$0.010

12Claude Sonnet 5+3 altsAnthropic28.1$2.0/M$10.0/M$0.214

13Claude Opus 4.6+1 altAnthropic27.1$5.0/M$25.0/M$0.554

14Claude Haiku 4.5NEWAnthropic19.8$1.0/M$5.0/M$0.152

◆ solid = independently verified · aggregator attested · vendor attributed — outlined = cross-reference · source-attributed. Dimmed ◆ = confidence inferred. “+N alts” = alternate disclosures in the drawer with why-demoted. Prices per million tokens; $/point = input price ÷ score. Every row opens its model dossier.

Verification: 14 models scored · 0 independently verified · 14 vendor-reported · 0 with source disagreement. How these tiers are assigned