[KongchangAI]
Expired50% confidenceEventExact time

The UP host's self-built benchmark has been updated with scores for the Claude series, GPT-5.6, and Claude 5.1, and plans to incorporate DeepSeek's new models and GPT-6.

1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/11/2026
First Seen

This claim has expired. It is kept for the record and no longer describes the current state.

Valid until: 9/25/2026(expired)

Sources

DeepSeek V4.1 Flash Early Access: Two Real-World Test Cases Put It to the Test

Related Entities

Claude 5.1modelGPT-5.6modelGPT-6model

Related Claims

UnverifiedGPT-5.6 adopts a naming strategy of upgrading iteratively under a fixed name, similar to the approach Anthropic uses when naming Claude.74% similar

Cite This Claim

Stable URI
https://kongchang.com/claim/908650
API
curl https://kongchang.com/api/v1/knowledge/claims/908650
MCP
get_claim(id=908650)