Compare

Gemini 2.5 Flash vs GPT-5 Turbo

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

Verdict

GPT-5 Turbo leads on LMSYS ELO (1398 vs 1312). Gemini 2.5 Flash is ~20.0× cheaper on a 3:1 input:output blend.

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

OpenAI

OpenAI’s flagship unified model. Handles text, vision, and audio natively. The generalist benchmark champion.

Native multimodalStrong math and scienceHuge ecosystem

Pricing

Input / 1M$0.30$6.00

Output / 1M$1.20$24.00

Context1.0M400K

Max output32K32K

LicenseProprietaryProprietary

Released2026-02-052026-01-21

Benchmarks

LMSYS ELO1312.01398.0

MMLU Pro82.191.2

HumanEval85.893.0

SWE-Bench—65.8

MATH84.291.5

GPQA—68.1

MMMU74.678.8

IFEval86.188.9

Per-task cost

Summarize a 1-hour meeting transcript$0.0051$0.102

Review a 500-line pull request$0.0048$0.096

Answer a customer support ticket$0.0019$0.037

Extract structured data from a resume$0.0027$0.054

Debug a stack trace with context$0.0042$0.084

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

Add up to four models, tweak tasks live.