24 models tracked · known for Claude 3 Opus, Claude 3 Sonnet, Claude 3.5 Haiku
Claims drawn from cited facts, not live model generation.
Anthropic is a lab whose blog carries its latest newsroom updates.
3 cited facts
Anthropic's tracked footprint in this graph consists of 24 shipped models and 255 recent blog announcements. With no papers count tracked, this mix—strong blog activity alongside a substantial model count—reads as an actively communicating lab that is still clearly investing in shipping models.
4 cited facts
382 benchmark scores across this lab’s models, tagged by evidence status.
| Model | Best score | Claim status |
|---|---|---|
| Claude Fable 5 | 100 | independently reproduced |
| Claude Opus 5 | 99 | independently reproduced |
| Claude Opus 4.8 | 98.33 | independently reproduced |
| Claude Opus 4.7 | 97.8 | independently reproduced |
| Claude Sonnet 4.5 | 97.73 | independently reproduced |
| Claude Haiku 4.5 | 96.36 | independently reproduced |
| Claude Sonnet 5 | 94.72 | independently reproduced |
| Claude Opus 4.6 | 94.44 | independently reproduced |
| Claude 3.7 Sonnet | 91.16 | independently reproduced |
| Claude 2 | 87.5 | independently reproduced |
| Claude Instant | 86.7 | self-reported |
| Claude Sonnet 4.6 | 86.5 | independently reproduced |
| Claude Opus 4.5 | 86.1 | independently reproduced |
| Claude Opus 4 | 85.05 | independently reproduced |
| Claude Opus 4.1 | 84.7 | independently reproduced |
| Claude Sonnet 4 | 84.37 | independently reproduced |
| Claude 3.5 Sonnet (October 2024) | 83.07 | independently reproduced |
| Claude 3.5 Sonnet | 82 | independently reproduced |
| Claude 3 Opus | 79.47 | independently reproduced |
| Claude 3.5 Haiku | 73.5 | independently reproduced |
| Claude 3 Sonnet | 67.87 | independently reproduced |
| Claude 3 Haiku | 65.07 | independently reproduced |
| Claude 2.1 | 64.67 | independently reproduced |
Where a model from Anthropic holds the top score, with the benchmark’s integrity grade.
Source facts, citations, and refresh stamp for this record.
Sources: curated_models, epoch_benchmarks, lab_blog_rss