Back to GalleryBack

Model Comparison

sonar-reasoningvsclaude-opus-4

Model Comparison

Compare performance, benchmarks, and characteristics

sonar-reasoning

openrouter

Context127K tokens

Input Price$1 / 1M tokens

Output Price$5 / 1M tokens

View Details →

claude-opus-4

anthropic

Context200K tokens

Input Price$15 / 1M tokens

Output Price$75 / 1M tokens

View Details →

Loading comparison...

Model Comparison

Metric	sonar-reasoning openrouter	claude-opus-4 anthropic
Pricing
Input Price	$1 / 1M tokens	$15 / 1M tokens
Output Price	$5 / 1M tokens	$75 / 1M tokens
Capabilities
Context Window	127K tokens	200K tokens
Capabilities		tools
Input type	text	text, image
Category Scores
Overall Average	56.6	55.9
Academia	48.3	70.1
Finance	59.4	N/A
Marketing	34.2	43.5
Maths	84.6	N/A
Science	62.3	80.9
Writing	51.1	32.0
Programming	N/A	53.1
Benchmark Tests
AIME	77.0	_
AA Coding Index	_	53.1
AAII	34.2	59.3
AA Math Index	84.6	_
GPQA	62.3	80.9
HLE	_	11.9
LiveCodeBench	_	65.4
MATH-500	92.1	_
MMLU-Pro	_	88.0
SciCode	_	40.9

sonar-reasoning Advantages

• Higher benchmark scores (56.6 vs 55.9)
• Lower input token cost ($1 / 1M tokens vs $15 / 1M tokens)

claude-opus-4 Advantages

• Larger context window (200K tokens vs 127K tokens)

Compare openrouter/sonar-reasoning vs anthropic/claude-opus-4