E-GLOBAL-001accepted
Opus 4.8 SWE-Bench Pro 69.2% vs GPT-5.5 58.6%
Anthropic launch / DataCamp comparison, May 2026
benchmark / high confidenceOCG LabTask-Domain Decision System
Evidence Ledger
Global Evidence
Global benchmark and claim records from the OCG Bench content master.
E-GLOBAL-002accepted
GPT-5.5 leads Terminal-Bench 2.1: 78.2% vs Opus 4.8 74.6%
Anthropic launch table, May 2026
benchmark / high confidenceE-GLOBAL-003bounded
Opus 4.8 hallucination rate 35.9% vs GPT-5.5 86% on AA-Omniscience
CodingFleet benchmark analysis, May 2026
third_party_test / medium confidencesingle third-party source; not independently verifiedE-GLOBAL-004bounded
Opus 4.8 is 4x less likely than Opus 4.7 to miss code flaws; 17x less likely than Sonnet 4.6 to produce dishonest summaries
Anthropic release notes, May 2026
vendor_claim / medium confidenceE-GLOBAL-005bounded
Sonnet 4.6 matches Opus 4.6 on OfficeQA enterprise document reading
Anthropic Sonnet 4.6 release, 2026
vendor_claim / medium confidenceE-GLOBAL-006bounded
Opus 4.8 Finance Agent v2: highest recorded score
Anthropic launch, May 2026
vendor_claim / medium confidenceE-GLOBAL-007accepted
Opus 4.8 OSWorld-Verified 83.4% vs GPT-5.5 78.7%
Anthropic launch table, May 2026
benchmark / high confidenceE-GLOBAL-008bounded
Opus 4.8 Legal Agent Benchmark: highest recorded, first to break 10% all-pass
Anthropic launch, May 2026
vendor_claim / medium confidenceE-GLOBAL-009bounded
Sonnet 4.6 vs GPT-5.5: Sonnet leads on coding; GPT-5.5 leads aggregate (89 vs 82)
BenchLM.ai comparison, June 2026
third_party_test / medium confidenceE-GLOBAL-010bounded
Opus 4.8 on 8-needle 1M MRCR v2: 76%; Sonnet 4.5 baseline: 18.5%
Anthropic Opus 4.6 release, Feb 2026
benchmark / medium confidence4.6 generation data; 4.8 not separately publishedE-GLOBAL-011bounded
Harvey BigLaw Bench: Opus 4.6 scored 90.2%
The Legal Prompts, Feb 2026
third_party_test / medium confidence