Model Comparison

ernie-4.5-21b-a3b-thinkingvsclaude-opus-4.6

Compare performance, benchmarks, and characteristics

openrouter

Context131072 tokens

Input Price$0.07 / 1M tokens

Output Price$0.28 / 1M tokens

anthropic

Context200K tokens

Input Price$5 / 1M tokens

Output Price$25 / 1M tokens

Model Comparison

Metric	ernie-4.5-21b-a3b-thinking openrouter	claude-opus-4.6 anthropic
Pricing
Input Price	$0.07 / 1M tokens	$5 / 1M tokens
Output Price	$0.28 / 1M tokens	$25 / 1M tokens
Capabilities
Context Window	131072 tokens	200K tokens
Capabilities	reasoning	tools, reasoning
Input type	text	text, image
Category Scores
Overall Average	40.4	60.6
Maths	41.3	N/A
Finance	28.1	N/A
Science	71.1	78.1
Writing	42.3	55.0
Academia	48.0	68.8
Marketing	37.4	53.4
Programming	14.5	47.6
Benchmark Tests
HLE	3.5	18.6
AIME	49.3	_
DROP	28.6	_
GPQA	81.1	91.3
MMLU	41.9	_
SciCode	31.5	45.7
MATH-500	93.1	_
MMLU-Pro	77.6	_
LiveCodeBench	46.7	_
AA Math Index	41.3	_
AA Coding Index	14.5	47.6
AAII	14.9	46.4

ernie-4.5-21b-a3b-thinking Advantages

claude-opus-4.6 Advantages

Loading comparison...

Model Comparison

Metric	ernie-4.5-21b-a3b-thinking openrouter	claude-opus-4.6 anthropic
Pricing
Input Price	$0.07 / 1M tokens	$5 / 1M tokens
Output Price	$0.28 / 1M tokens	$25 / 1M tokens
Capabilities
Context Window	131072 tokens	200K tokens
Capabilities	reasoning	tools, reasoning
Input type	text	text, image
Category Scores
Overall Average	40.4	60.6
Maths	41.3	N/A
Finance	28.1	N/A
Science	71.1	78.1
Writing	42.3	55.0
Academia	48.0	68.8
Marketing	37.4	53.4
Programming	14.5	47.6
Benchmark Tests
HLE	3.5	18.6
AIME	49.3	_
DROP	28.6	_
GPQA	81.1	91.3
MMLU	41.9	_
SciCode	31.5	45.7
MATH-500	93.1	_
MMLU-Pro	77.6	_
LiveCodeBench	46.7	_
AA Math Index	41.3	_
AA Coding Index	14.5	47.6
AAII	14.9	46.4

ernie-4.5-21b-a3b-thinking Advantages

claude-opus-4.6 Advantages