Unverified50% confidenceEventExact time
Using Claude Max mode consumed 2% of a tester's monthly quota from a single query asking 'what model are you?'
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Claude Opus 4.8 Hands-On: What Can You Build in One Hour?
bilibili神烦老狗5/29/2026
Related Claims
UnverifiedClaude Opus 4.8 的 Max 模式额度消耗极高，测试者仅问一句'你是什么模型'就消耗了 2% 的月度额度75% similarUnverifiedQwen 3.6 Max中性模式下歧义检测F1仅16%，平均每任务只问0.07个问题，但一旦提问澄清准确率高达94.7%56% similarUnverifiedMiniMax M2.7每任务提问0.6个问题，但澄清推进率只有60.7%，存在低信息增益提问问题55% similarUnverified本次评测使用全国一卷高考数学题对比Muse Glimmer和通义千问3.6 27B54% similarUnverified千问三235B MoE总参数量2350亿，但每次推理实际激活的参数约22B54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/58792API
curl https://kongchang.com/api/v1/knowledge/claims/58792MCP
get_claim(id=58792)