Compare

o4 vs Gemini 2.5 Flash

OpenAI vs Google DeepMind. Specs, benchmarks, and real per-task cost — all in one page.

Verdict

Gemini 2.5 Flash is ~50.0× cheaper on a 3:1 input:output blend.

OpenAI

Deep-reasoning model. Spends more tokens thinking to crush math, science, and hard coding problems.

State-of-the-art math and scienceExtremely high-ceiling reasoning

Google DeepMind

Google’s price/performance darling. 1M context at $0.30/M input — nobody comes close on throughput-per-dollar.

Absurd cost-to-context ratioFast multimodalFree tier via AI Studio

Pricing

Input / 1M$15.00$0.30

Output / 1M$60.00$1.20

Context200K1.0M

Max output100K32K

LicenseProprietaryProprietary

Released2026-02-182026-02-05

Benchmarks

LMSYS ELO—1312.0

MMLU Pro93.882.1

HumanEval92.385.8

SWE-Bench68.0—

MATH96.284.2

GPQA82.5—

MMMU—74.6

IFEval—86.1

Per-task cost

Summarize a 1-hour meeting transcript$0.255$0.0051

Review a 500-line pull request$0.240$0.0048

Answer a customer support ticket$0.093$0.0019

Extract structured data from a resume$0.135$0.0027

Debug a stack trace with context$0.210$0.0042

Per-call cost using published token counts for each task. Real-world prompts vary.

Add up to four models, tweak tasks live.