#14Anthropic· released 2026-06-30
Anthropic: Claude Sonnet 5.
Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
View on OpenRouter ↗Context
1.0M
Input $/1M
$2.00
Output $/1M
$10.00
Modality
text · image · file
Benchmarks
How these are aggregated →Arena Elo
1503
LMArena — human blind votes
AA Index
59.0
Artificial Analysis composite
LiveBench
76.2
Contamination-resistant
GPQA
83.4%
Grad-level science reasoning
MMLU-Pro
82.1%
Broad knowledge
SWE-bench
67.2%
Real GitHub bugs (Verified)
Terminal-Bench
44.9%
Agentic shell tasks
ARC-AGI
14.6%
Abstract reasoning
Aider Polyglot
82.4%
Multi-language code editing
Speed
64 tok/s
Median output speed
Nearby on the leaderboard
- Elo 1588
- Elo 1571
- Elo 1565
- Elo 1556