Rankings / Anthropic
Claude 2
released 2023-07-11
BenchAtlas Index
as of 2026-07-14
base configuration
rank #394
4 families · 3 categories · medium
Benchmark evidence
7 results
coding
| LiveCodeBench v6 implementation=artificial-analysis | 17.1 % | independent | Artificial Analysis | |
| SciCode | 19.4 % | independent | Artificial Analysis |
external indices
| AA Coding Index | 12.9 points | independent | Artificial Analysis | |
| Intelligence Index v4.1 | 3.6 points | independent | Artificial Analysis |
knowledge science
| GPQA Diamond implementation=artificial-analysis | 34.4 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=artificial-analysis | 48.6 % | independent | Artificial Analysis |
reasoning math
| AIME implementation=artificial-analysis | 0.0 % | independent | Artificial Analysis |
Agent + model results
systems, not bare-model scores
| agent + model RAG + Claude 2 | SWE-bench Verified | 4.4 % | community | SWE-bench Leaderboard |
| agent + model RAG + Claude 2 | SWE-bench Lite | 3.0 % | community | SWE-bench Leaderboard |
| agent + model RAG + Claude 2 | SWE-bench (full test split) | 2.0 % | community | SWE-bench Leaderboard |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.