Compare deepinfra/qwen-2.5-72b-instruct vs gemini/gemini-3-flash-preview

Model Comparison

Metric	qwen-2.5-72b-instruct deepinfra	gemini-3-flash-preview gemini
Pricing
Input Price	$0.12 / 1M tokens	$0.5 / 1M tokens
Output Price	$0.39 / 1M tokens	$3 / 1M tokens
Capabilities
Context Window	32768 tokens	1M tokens
Capabilities	tools	tools, reasoning
Input type	text	text, image, audio, video
Category Scores
Overall Average	32.5	0.0
Maths	14.0	N/A
Finance	21.5	N/A
Science	52.8	N/A
Writing	40.0	N/A
Academia	39.0	N/A
Marketing	40.4	N/A
Programming	19.5	N/A
Benchmark Tests
HLE	4.2	_
AIME	16.0	_
GPQA	49.0	_
SciCode	26.7	_
MATH-500	85.8	_
MMLU-Pro	71.5	_
HumanEval	86.6	_
LiveCodeBench	27.6	_
AA Math Index	14.0	_
AA Coding Index	19.5	_
AAII	29.0	_

qwen-2.5-72b-instruct Advantages

• Higher benchmark scores (32.5 vs 0.0)
• Lower input token cost ($0.12 / 1M tokens vs $0.5 / 1M tokens)

gemini-3-flash-preview Advantages

• Larger context window (1M tokens vs 32768 tokens)

Compare deepinfra/qwen-2.5-72b-instruct vs gemini/gemini-3-flash-preview