Gemini 3.1 Flash-Lite
GoogleGeminiProprietary
Описание
Gemini 3.1 Flash-Lite is the first Flash-Lite model in the Gemini 3 series. It is optimized for high-volume, latency-sensitive tasks like translation, content moderation, and classification. It delivers enhanced performance at a fraction of the cost of larger models, with 2.5x faster Time to First Answer Token and 45% increased output speed compared to 2.5 Flash. Supports text, image, video, audio, and PDF input with a 1 million-token context window.
Дата выхода
2026-03-03
Параметры
—
Длина контекста
1.0M
Модальности
audio, image, pdf, text, video
Радар способностей
16
general
36
coding
82
reasoning
56
science
10
agents
80
multimodal
Рейтинги
| Домен | #Место | Оценка | Источник |
|---|---|---|---|
| Агентные возможности | 90 | 37.0 | LS |
| Рейтинг кодинга | 290 | 52.0 | AA |
| Общий рейтинг | 308 | 42.0 | AA |
| Мультимодальный рейтинг | 46 | 58.0 | LS |
| Наука | 222 | 57.0 | AA |
Оценки бенчмарков (LLM Stats)
(LLM Stats (zeroeval))Agents
Finance Agent v2
30.0%
Factuality
SimpleQA
43.3%Сам.
Language
MMMLU
88.9%Сам.
Legal
Legal Agent Benchmark
0.0%
Long Context
MRCR v2 (8-needle)
60.1%Сам.
Multimodal
VideoMMMU
84.8%Сам.
Reasoning
GPQANYU + Cohere + Anthropic (2023)
86.9%Сам.
CharXiv-R
73.2%Сам.
FACTS Grounding
40.6%Сам.
Humanity's Last Exam
16.0%Сам.
Vision
MMMU-Pro
76.8%Сам.
Индексы оценки AA
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))82.2
Ifbench(Google Research (2023))77.2
Lcr(Artificial Analysis)74.3
Scicode(UIUC + Argonne National Lab (2024))43.4
Coding Index(Artificial Analysis)34.7
Tau2(Sierra + U Toronto + Vector Institute (2025))31.3
Terminalbench V2 131.1
Terminalbench Hard(Stanford × Laude Institute (2026))24.2
Hle(Center for AI Safety + Scale AI (2025))17.2
Intelligence Index(Artificial Analysis)15.6
Tau Banking9.7
Terminalbench V4 00.5
Оценки категорий LLM Stats
(LLM Stats (zeroeval))Language90
Physics90
Biology90
Chemistry90
Multimodal80
Reasoning60
Vision60
Math50
General50
Healthcare50
Long Context40
Factuality40
Grounding40
Finance30
Agents10
Legal0
Цены
Цена ввода$0.25 / 1M токенов
Цена вывода$1.5 / 1M токенов
Смешанная цена (3:1)$0.563 / 1M токенов
Цена чтения кэша$0.025 / 1M токенов
Скорость
Токенов/сек0.0
Задержка первого токена0.00s
Время до первого ответа0.00s
Рейтинг цен провайдеров
Рейтинг цен провайдеров
25 провайдеров
Самый дешевый: GoogleСамый дорогой: Cortecs
ПровайдерВводВывод
1GoogleСамый дешевый
$0
$0
2DeepInfra
$0
$0
3Kilo Gateway
$0.125
$0.75
4302.AI
$0.25
$1.5
5NanoGPT
$0.25
$1.5
6Abacus
$0.25
$1.5
7OpenRouter
$0.25
$1.5
8ZenMux
$0.25
$1.5
9Vivgrid
$0.25
$1.5
10SAP AI Core
$0.25
$1.5
11Poe
$0.25
$1.5
12AIHubMix
$0.25
$1.5
13Requesty
$0.25
$1.5
14Vercel AI Gateway
$0.25
$1.5
15DevPass (LLM Gateway)
$0.25
$1.5
16Vertex
$0.25
$1.5
17NEAR AI Cloud
$0.25
$1.5
18OrcaRouter
$0.25
$1.5
19Merge Gateway
$0.25
$1.5
20Pioneer
$0.25
$1.5
21Ofox
$0.25
$1.5
22Impossibl
$0.25
$1.5
23Eden AI
$0.25
$1.5
24Tempr Gateway
$0.25
$1.5
25Cortecs
$0.272
$1.631
Сравнение цен разных API-провайдеров для этой модели.