Expired50% confidenceBenchmarkExact time
In the Cupcake Agent real-world test, GPT-5.6 Luna had the lowest latency but higher token consumption, consuming approximately 47,000 tokens across 7 requests, making it a token-intensive model.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/28/2026
First Seen
Valid until: 10/12/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedThe author's cumulative use of GPT-5.6 Luna plus GPT-6 Luna consumed approximately 500 million tokens over three to four days, with only 3% of the quota deducted.74% similarVerifiedOn equivalent tasks, GPT-5.5 uses approximately 72% fewer output tokens than Opus.74% similarUnverifiedCompared to GPT-5.5, GPT-5.6 consumes approximately one-third of the Tokens per PR, and the median completion time is nearly halved.73% similarUnverifiedGPT-5.6 Luna has a quality index of approximately 0.7875, a latency of approximately 24 seconds, and an attack rate of 2.8.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/955247API
curl https://kongchang.com/api/v1/knowledge/claims/955247MCP
get_claim(id=955247)