Gemini 3.7 Flash (high)
GoogleGeminiProprietary
描述
Google's most intelligent Flash workhorse yet for coding and agents. Ships ~3 weeks after Gemini 3.6 Flash with gains on debugging/issue resolution, first-pass code accuracy, FrontierCode 1.1 Main (43.6% vs 3.6's 34.4%), and DeepSWE v1.1 (65.3% vs 49.0%). WebDev Arena Elo 1588 vs 1538; GDP.pdf 34.0% vs 22.0%; AutomationBench 30.4% vs 17.0%. Text/image/video/audio/PDF in, text out; 1M context / 65,536 max output. Powers Gemini Spark for Pro/Ultra. Intro pricing is half of original 3.6 Flash ($0.75/$3.75 per 1M input/output tokens) through 2026-12-31; list price $1.50/$7.50 applies starting 2027-01-01.
發布日期
2026-08-13
參數規模
—
上下文長度
1.0M
支援模態
audio, image, pdf, text, video
能力雷達圖
54
general
73
coding
95
reasoning
72
science
50
agents
80
multimodal
排行榜排名
基準測試分數 (LLM Stats)
(LLM Stats (zeroeval))Agents
GDPval-AA
1525.00 / 3000自報
Terminal-Bench 2.1
85.8%自報
DeepSWE 1.1
65.3%自報
OSWorld 2.0
47.9%自報
FrontierCode 1.1
43.6%自報
AutomationBench
30.4%自報
Agents' Last Exam
26.3%自報
Terminal-Bench 3.0
14.9%自報
Biology
BioMysteryBench
43.5%自報
Code
WebDev Arena
1588.00 / 2000自報
General
MRCR v2 (8-needle)
97.0%自報
Artificial Analysis
56.0%自報
GDP.pdf
34.0%自報
Long Context
LVBench
85.4%自報
Multimodal
CharXiv-R
88.7%自報
AA 評測指數
(Artificial Analysis)Coding Index(Artificial Analysis)76.1
Intelligence Index(Artificial Analysis)56.0
Gpqa(NYU + Cohere + Anthropic (2023))0.9
Terminalbench V2 10.9
Lcr(Artificial Analysis)0.8
Scicode(UIUC + Argonne National Lab (2024))0.6
Hle(Center for AI Safety + Scale AI (2025))0.5
Tau Banking0.3
LLM Stats 分類評分
(LLM Stats (zeroeval))Legal100
Finance100
Agents100
Reasoning100
General100
Long Context90
Multimodal60
Code60
Vision60
Tool Calling50
Science40
Biology40
定價
輸入價格$0.75 / 1M tokens
輸出價格$3.75 / 1M tokens
混合價格(3:1)$1.5 / 1M tokens
快取讀取價格$0.075 / 1M tokens
速度
Tokens/秒515.0
首Token延遲6.35s
首回答延遲6.35s
供應商價格排行
供應商價格排行
1 個供應商
供應商輸入輸出
1Google
$0
$0
比較該模型在不同 API 供應商之間的定價。