← All models

Gemini · Last 7 days

Gemini 3.7 Flash

0–100 · higher means more positive community experience, not benchmark performance.

Updated 2026-09-05 14:21 UTC · 2026-08-29 – 2026-09-05 UTC
Community score46.8/10032 comments in 7 days

Community reviews

Selected comments from the last 7 days. The balance of excerpts does not represent the share of positive reviews.

17 selected excerpts

NegativeCoding

“gemini flash 3.7 42/100”

PositiveGeneral text

“3.7 is still okay for simple daily office work, the experience of writing webpages is also decent, mainly it's really cheap”

ZhihuzhMachine translated
Show original

3.7日常简单办公还可以,写网页的体验也还行,主要是真便宜

PositiveGeneral text

“The writing feels more nimble than Gemini 3.7 Flash, without that stuffy feeling, and it can write jokes that make people laugh”

ZhihuzhMachine translated
Show original

写作感觉要比 Gemini 3.7 Flash 更灵动一点,没有那种闷闷的感觉,俏皮话写得也是能博人一笑的。

PositiveGeneral text

“I did the whole itinerary with Opus first, but when I gave it to Gemini 3.7 Flash to review, it started correcting it with "this place will close 5PM" or "this place is closed for good". It was right on every nit”

PositiveReasoning

“After 3.7 Flash it's basically back to T1”

RedNotezhMachine translated
Show original

3.7 Flash之后基本回到T1了

PositiveSpeed & latency

“Gemini 3.7 Flash is already outperforming top tier Pro models from 6 months ago. Fast optimized models are the future.”

NeutralCoding

“execute with Gemini 3.7 flash. It can work well especially if you have Claude split the work in to small tickets/ tasks then have a 3.7 agent work each ticket”

NegativeReasoning

“Gemini 3.7 Flash, Composer 2.5 → Speed. Hallucinations. Impreciseness. They are cheap and fast, perfect for small UI and text changes or small corrections, but if you give them task that's more complicated or requires taste, expect slop.”

NegativeReasoning

“Gemini 3.7 Flash is still very LLM-y as it often doesn't seem to the take the given context into consideration and consistently hallucinates”

NeutralGeneral text

“The one that asks questions at the beginning is Gemini Flash 3.7.”

PositiveReasoning

“I found gemini 3.7 flash very proactive, actually. It infered things correctly that I ever thought to specified, and went the extra mile”

PositiveReasoning

“World knowledge is also strong.”

ZhihuzhMachine translated
Show original

世界知识也强

NegativeSpeed & latency

“meanwhile gemini flash 3.7 is not faster, no idea what is going on. it takes endless with minimum context.. and my connection is great.”

NegativeSpeed & latency

“gemini 3.7 flash with google ai studio API is ~130 token/s, google vertex API is ~80 token/s, who is going to make up the 3-6x speed difference I need”

ZhihuzhMachine translated
Show original

gemini 3.7 flash用google ai studio的API在~130 token/s,google vertex的API在~80 token/s,我缺的3-6倍速度谁给我补上

PositiveSpeed & latency

“cheap and fast on the 20$ plan and provides way more generous usage than ChatGPT”

NegativeSpeed & latency

“Gemini 3.7 flash outputs so many tokens per answer it doesn't matter how fast its TPS is, sol will end up being both cheaper and faster than Gemini”

NegativeRoleplay / creative

“A professional screenwriter told me that Gemini 3.7 Flash is really not useful for creative assistance, can only be used like Baidu.”

ZhihuzhMachine translated
Show original

专业编剧告诉我,Gemini 3.7 Flash在创作辅助上是真没什么用,只能当百度来用。