Home/Use Cases/Math & Science

Best LLM for Math & Science

3 models ranked for math & science tasks. Sorted by benchmark quality score, with price as a secondary factor.

Best Quality
o3
OpenAI
ELO 1395
Cheapest Option
DeepSeek V3.2 (Reasoner)
DeepSeek
$0.28/1M/1M input

All Models for Math & Science

#ModelProviderInput / 1MOutput / 1MELOFlags
๐Ÿฅ‡o3OpenAI$2.00/1M$8.00/1M1395
๐ŸฅˆDeepSeek V3.2 (Reasoner)DeepSeek$0.28/1M$0.42/1M1385
OSS
๐Ÿฅ‰o4-miniOpenAI$1.10/1M$4.40/1M1350

Why We Picked These Models

o3
$2.00/1M/1MELO 1395

o3 is OpenAI's mainstream reasoning model, replacing the original o1 with a dramatic price cut and better benchmark scores. Top-tier reasoning on math, science, and agentic tasks.

DeepSeek V3.2 (Reasoner)
$0.28/1M/1MELO 1385

DeepSeek V3. o-series-class reasoning at a fraction of the price.

o4-mini
$1.10/1M/1MELO 1350

o4-mini is OpenAI's most cost-effective reasoning model, offering o-series chain-of-thought capability at roughly half the price of o3. Cheapest reasoning model in the OpenAI lineup.

Compare Top Models

o3 vs DeepSeek V3.2 (Reasoner)o3 vs o4-miniDeepSeek V3.2 (Reasoner) vs o4-mini

Frequently Asked Questions

What is the best LLM for math & science?

o3 by OpenAI is rated as the best model for math & science with an ELO score of 1395. Top-tier reasoning on math, science, and agentic tasks.

What is the cheapest LLM for math & science?

DeepSeek V3.2 (Reasoner) is the most affordable option for math & science at $0.28/1M per 1M input tokens.

Is there a free LLM for math & science?

No completely free models are listed for math & science, but DeepSeek V3.2 (Reasoner) start at very low prices.