Google
Gemini 3.8 Flash (Medium Reasoning)
Balanced reasoning effort variant of Gemini 3.8 Flash for agentic workflows.
Intelligence
65.2 #190
Top 27% tier
Output speed
315 t/s
TTFT 160 ms · ITL 3.2 ms/tok
Blended price (3:1)
$1.5 /1M
In $0.75/M · Out $3.75/M
Context
1M
≈ 2,500 A4 pages · max out 65536
Quality Domain Radar
6-Axis Frontier Evaluation vs Category Average & Flagship
Gemini 3.8 Flash (Medium Reasoning)
Claude Mythos Preview (#1)
Average (68)
Radar axes are approximated from category scores where direct benchmark measurements are missing — see Методология.
API deployment
Compare hosts & latency
Throughput, TTFT and cache pricing across Groq, Cerebras, Together, DeepInfra, Fireworks, Bedrock and Azure.
GroqCerebrasTogether AIDeepInfraFireworks AIAWS BedrockAzure AI
Groq
Cerebras
Together AI
DeepInfra
Fireworks AI
AWS Bedrock
Azure AIОценки по источникам
Оценки по категориям
Результаты бенчмарков
Детали модели
Дата выпуска
1 сент. 2026 г.
Параметры
Not disclosed
Лицензия
Proprietary
Контекстное окно
1,000,000 tokens
Цена ввода
$0.75/M
Цена вывода
$3.75/M
Скорость вывода
315 t/s
Время до первого токена
160 ms
Estimated monthly bill
Estimated monthly spend
$0.00
Calculating savings vs GPT-4o...
Cite & embed
GitHub Markdown badge
[](https://llmpodium.com/models/gemini-3-8-flash-medium)BibTeX
@misc{llmpodium2026gemini38flashmedium,
title = {Benchmark Analysis and Intelligence Evaluation of Gemini 3.8 Flash (Medium Reasoning)},
author = {LLMPodium Research},
year = {2026},
url = {https://llmpodium.com/models/gemini-3-8-flash-medium}
}More from Google
Альтернативные модели
Ближайшие альтернативы для Gemini 3.8 Flash (Medium Reasoning) по возможностям, цене и бенчмаркам.
