readable AI benchmarks

Anthropic

Claude Sonnet 5 (Non-reasoning, High Effort)

non-reasoning · closed weights · Jun 30, 2026

Quality Score51.2
Value Score34.0
Reliability49.7
Cache Discount90%

Model specification

Reasoning
non-reasoning
Input modalities
text, image
Output modalities
text
Context window
1M tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$2.00
Output
$10.00
Cached Input
$0.200
Cached Output
$2.50
Cost per task
$0.417

Capability

Intelligence
42.6
Coding
66.4
Agentic
34.2
Omniscience
-0.7
Correct
33.8%
Blended price
$1.54

Answer outcomes

Correct 33.8%Incorrect 34.5%Abstained 31.7%

Artificial Analysis benchmarks

GDPval-AA v2
44%
τ³-Banking
16%
Terminal-Bench v2.1
75%
SciCode
49%
Humanity’s Last Exam
19%
GPQA Diamond
80%
CritPt
1%
AA-Omniscience
50%
AA-LCR
66%

Similar models