Claude 4 Sonnet 20250514

release date unknown

BenchAtlas Index

as of 2026-07-14

Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

0 results

No bare-model results yet.

Agent + model results

systems, not bare-model scores

agent + model Lingxi-v1.5 + Claude 4 Sonnet 20250514	SWE-bench Verified	74.6 %	unverified	SWE-bench Leaderboard
agent + model Tools + Claude 4 Sonnet 20250514	SWE-bench Verified	72.4 %	unverified	SWE-bench Leaderboard
agent + model Bloop + Claude 4 Sonnet 20250514	SWE-bench Verified	71.2 %	unverified	SWE-bench Leaderboard
agent + model Moatless Tools + Claude 4 Sonnet 20250514	SWE-bench Verified	70.8 %	community	SWE-bench Leaderboard
agent + model OpenHands + Claude 4 Sonnet 20250514	SWE-bench Verified	70.4 %	community	SWE-bench Leaderboard
agent + model SWE-agent + Claude 4 Sonnet 20250514	SWE-bench Verified	66.6 %	community	SWE-bench Leaderboard
agent + model mini-SWE-agent + Claude 4 Sonnet 20250514	SWE-bench Verified	64.9 %	community	SWE-bench Leaderboard
agent + model mini-SWE-agent + Claude 4 Sonnet 20250514	SWE-bench bash-only	64.9 %	community	SWE-bench Leaderboard
agent + model KGCompass + Claude 4 Sonnet 20250514	SWE-bench Lite	58.3 %	unverified	SWE-bench Leaderboard
agent + model SWE-agent + Claude 4 Sonnet 20250514	SWE-bench Lite	56.7 %	community	SWE-bench Leaderboard
agent + model OpenHands-Versa + Claude 4 Sonnet 20250514	SWE-bench Multimodal	34.4 %	community	SWE-bench Leaderboard
agent + model mini-SWE-agent + Claude 4 Sonnet 20250514	SWE-bench bash-only	0.4 usd_per_task	community	SWE-bench Leaderboard
agent + model mini-SWE-agent + Claude 4 Sonnet 20250514	SWE-bench Verified	0.4 usd_per_task	community	SWE-bench Leaderboard

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.

Related variants

Claude 4 Sonnet