Frontier Log

AnthropicModel

Claude Sonnet 5

Made by
Anthropic
Released
Announced in
Introducing Claude Sonnet 5anthropic.com
Event
Claude Sonnet 5
API record
/api/v1/models/claude-sonnet-5

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools43.2%
CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
reported later, in Claude Sonnet 5.5 System Card43.1%
CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Humanity's Last ExamWith tools57.4%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
reported later, in Introducing Claude Sonnet 5.554.9%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
SWE-bench ProThe source doesn't say whether this is the public or the commercial set63.2%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 3 other lab documents: 1 2 3
BrowseCompadaptive thinking, max effort; 10M-token limit, compaction at 200K tokens84.7%
CheckedA person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
AutomationBenchZapier leaderboard, private held-out set; max effort13.5%
CheckedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
run by Zapier; private set, release 1.0.6 · reported later, in Claude Sonnet 5.5 System Card10.7%
CorroboratedThe same number is printed in 1 other lab document: 1
OSWorld 2.1Full set, partial score · max effort; mean of 5 runs; September 10, 2026 task files · reported later, in Introducing Claude Sonnet 5.557.0%
CorroboratedThe same number is printed in 1 other lab document: 1
OSWorld 2.1Full set, strict pass rate · max effort; mean of 5 runs; September 10, 2026 task files · reported later, in Claude Sonnet 5.5 System Card25.6%
CheckedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Sources

Reported by Anthropic

Events

Stories

All models