readable AI benchmarks

Anthropic

Claude Sonnet 4.6 (Non-reasoning, Low Effort)

non-reasoning · closed weights · Feb 17, 2026

Quality Score44.5
Value Score28.5
Reliability49.0
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
$3.00
Output
$15.00
Cached Input
$0.300
Cached Output
$3.75
Cost per task
not available

Capability

Intelligence
35.1
Coding
not available
Agentic
not available
Omniscience
-2.1
Correct
36.5%
Blended price
$2.31

Answer outcomes

Correct 36.5%Incorrect 38.6%Abstained 24.9%

Artificial Analysis benchmarks

SciCode
44%
Humanity’s Last Exam
11%
GPQA Diamond
80%
CritPt
1%
AA-Omniscience
49%
AA-LCR
65%

Similar models