Unverified50% confidenceFactTime unknown

Gemini 3.5 Flash consumed approximately 72 million tokens in Artificial Analysis's benchmarks, while GPT-5.5 Medium used only 22 million tokens.

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

7/2/2026

First Seen

Valid until: 9/30/2026

Sources

Gemini 3.5 Flash Falls Flat: Great Benchmarks, Terrible Real-World Performance, and a Buggy CLI

bilibili量子菠萝_

Related Claims

VerifiedGPT-3时代的上下文窗口仅有2,048个Token，而Google Gemini已支持百万级Token，Claude支持200K Token，GPT-4 Turbo支持128K Token77% similar UnverifiedGemini 3 Flash在最高思考级别下处理日常任务时，比2.5 Pro平均减少30%的Token消耗73% similar VerifiedGemini 3.1 Flash 和 Gemini 3.1 Pro 都支持 100 万 token 上下文窗口73% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/54509

API

curl https://kongchang.com/api/v1/knowledge/claims/54509

MCP

get_claim(id=54509)