Compare
Llama 4 Scout vs GPT-5 Turbo
Meta vs OpenAI. Specs, benchmarks, and real per-task cost — all in one page.
Verdict
GPT-5 Turbo leads on LMSYS ELO (1398 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.
Meta
Llama 4 ScoutThe 70B workhorse open-weight. Fits on a single H100, beloved by self-hosters.
Single-GPU deployableDirt cheap on aggregatorsStrong fine-tuning base
OpenAI
GPT-5 TurboOpenAI’s flagship unified model. Handles text, vision, and audio natively. The generalist benchmark champion.
Native multimodalStrong math and scienceHuge ecosystem
Pricing
Input / 1M$0.35$6.00
Output / 1M$0.80$24.00
Context128K400K
Max output8K32K
LicenseOpen weightsProprietary
Released2025-12-102026-01-21
Benchmarks
LMSYS ELO1265.01398.0
MMLU Pro77.291.2
HumanEval82.493.0
SWE-Bench—65.8
MATH72.591.5
GPQA—68.1
MMMU—78.8
IFEval84.188.9
Per-task cost
Summarize a 1-hour meeting transcript$0.0057$0.102
Review a 500-line pull request$0.0044$0.096
Answer a customer support ticket$0.0017$0.037
Extract structured data from a resume$0.0026$0.054
Debug a stack trace with context$0.0037$0.084
Per-call cost using published token counts for each task. Real-world prompts vary.