Compare anthropic/claude-opus-4 vs deepinfra/llama-guard-4-12b

Model Comparison

Metric	claude-opus-4 anthropic	llama-guard-4-12b deepinfra
Pricing
Input Price	$15 / 1M tokens	$0.18 / 1M tokens
Output Price	$75 / 1M tokens	$0.18 / 1M tokens
Capabilities
Context Window	200K tokens	163840 tokens
Capabilities	tools
Input type	text, image	text, image
Category Scores
Overall Average	62.5	0.0
Maths	73.3	N/A
Finance	63.8	N/A
Science	75.0	N/A
Writing	56.8	N/A
Academia	66.9	N/A
Marketing	57.4	N/A
Programming	44.2	N/A
Benchmark Tests
HLE	11.7	_
AIME	75.7	_
GPQA	79.6	_
SciCode	39.8	_
MATH-500	98.2	_
MMLU-Pro	87.3	_
LiveCodeBench	63.6	_
AA Math Index	73.3	_
AA Coding Index	44.2	_
AAII	54.2	_

claude-opus-4 Advantages

• Higher benchmark scores (62.5 vs 0.0)
• Larger context window (200K tokens vs 163840 tokens)

llama-guard-4-12b Advantages

• Lower input token cost ($0.18 / 1M tokens vs $15 / 1M tokens)

Compare anthropic/claude-opus-4 vs deepinfra/llama-guard-4-12b