Compare

Gemini 2.5 Ultra vs Llama 4 Scout

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

Verdict

Gemini 2.5 Ultra leads on LMSYS ELO (1385 vs 1265). Llama 4 Scout is ~22.7× cheaper on a 3:1 input:output blend. Llama 4 Scout is open-weights, the other is proprietary.

Google DeepMind

Gemini 2.5 Ultra

2M-token context, native video understanding, and Google’s deepest multimodal stack. The long-context king.

2M context windowNative videoBest multimodal reasoning

Meta

Llama 4 Scout

The 70B workhorse open-weight. Fits on a single H100, beloved by self-hosters.

Single-GPU deployableDirt cheap on aggregatorsStrong fine-tuning base

Pricing

Input / 1M$7.00$0.35

Output / 1M$21.00$0.80

Context2.0M128K

Max output64K8K

LicenseProprietaryOpen weights

Released2026-02-052025-12-10

Benchmarks

LMSYS ELO1385.01265.0

MMLU Pro90.477.2

HumanEval90.282.4

SWE-Bench58.3—

MATH92.072.5

MMMU82.1—

IFEval89.784.1

Per-task cost

Summarize a 1-hour meeting transcript$0.115$0.0057

Review a 500-line pull request$0.098$0.0044

Answer a customer support ticket$0.038$0.0017

Extract structured data from a resume$0.056$0.0026

Debug a stack trace with context$0.084$0.0037

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

Tool

Interactive compare →

Add up to four models, tweak tasks live.