Compare anthropic/claude-sonnet-4 vs bedrock/llama3-2-3b-instruct-v1.0

Model Comparison

Metric	claude-sonnet-4 anthropic	llama3-2-3b-instruct-v1.0 bedrock
Pricing
Input Price	$3 / 1M tokens	$0.15 / 1M tokens
Output Price	$15 / 1M tokens	$0.15 / 1M tokens
Capabilities
Context Window	200K tokens	128K tokens
Capabilities	tools
Input type	text, image	text, image
Category Scores
Overall Average	64.9	17.3
Maths	74.3	3.3
Finance	65.4	7.3
Science	72.7	30.5
Writing	63.5	21.2
Academia	67.1	22.0
Marketing	66.4	19.9
Programming	45.1	N/A
Benchmark Tests
HLE	9.6	5.2
AIME	77.3	6.7
GPQA	77.7	32.8
MMLU	_	63.4
SciCode	40.0	5.2
MATH-500	99.1	48.9
MMLU-Pro	84.2	34.7
LiveCodeBench	65.5	8.3
AA Math Index	74.3	3.3
AA Coding Index	45.1	_
AAII	56.5	11.2

claude-sonnet-4 Advantages

• Higher benchmark scores (64.9 vs 17.3)
• Larger context window (200K tokens vs 128K tokens)

llama3-2-3b-instruct-v1.0 Advantages

• Lower input token cost ($0.15 / 1M tokens vs $3 / 1M tokens)

Compare anthropic/claude-sonnet-4 vs bedrock/llama3-2-3b-instruct-v1.0