Compare deepinfra/phi-4 vs anthropic/claude-opus-4.5

Model Comparison

Metric	phi-4 deepinfra	claude-opus-4.5 anthropic
Pricing
Input Price	$0.07 / 1M tokens	$5 / 1M tokens
Output Price	$0.14 / 1M tokens	$25 / 1M tokens
Capabilities
Context Window	16384 tokens	200K tokens
Capabilities
Input type	text	text, image
Category Scores
Overall Average	70.3	76.6
Science	69.5	81.3
Writing	71.1	71.2
Maths	N/A	91.3
Finance	N/A	80.5
Academia	N/A	78.2
Marketing	N/A	73.3
Programming	N/A	60.2
Benchmark Tests
HLE	_	28.4
GPQA	65.8	86.6
SciCode	_	49.5
MMLU-Pro	74.3	89.5
LiveCodeBench	_	87.1
AA Math Index	_	91.3
AA Coding Index	_	60.2
AAII	_	69.8

phi-4 Advantages

• Lower input token cost ($0.07 / 1M tokens vs $5 / 1M tokens)

claude-opus-4.5 Advantages

• Higher benchmark scores (76.6 vs 70.3)
• Larger context window (200K tokens vs 16384 tokens)

Compare deepinfra/phi-4 vs anthropic/claude-opus-4.5