AI Model Timeline

Tracking the accelerating release frequency of frontier AI models.

Claude 3.5 Haiku

Anthropic·

LLMssmallcloud

Overview

Fastest model in its intelligence class.

Capabilities and innovations

Low latencyHigh intelligenceTool use

Benchmarks

BenchmarkScoreSource
GPQA Diamond61.8%unsourced
Humanity’s Last Exam (no tools)10.1%unsourced
SWE-bench Verified28.7%unsourced
MMLU85.9%unsourced
  • SWE-bench Pro: benchmark did not exist at release
  • MMLU-Pro: benchmark did not exist at release
  • Terminal-Bench 2.x: benchmark did not exist at release
  • MMMU-Pro: benchmark did not exist at release

Links

More from Anthropic

Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.