Unverified50% confidenceBenchmarkExact time
According to Artificial Analysis data, Haiku 5.5 consumes approximately 162 output tokens per intelligence index task in max effort mode, roughly 3 times that of GPT-6 Luna in the equivalent mode.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/8/2026
First Seen
Valid until: 1/6/2027
Sources
Related Entities
Related Claims
UnverifiedAccording to the author's analysis, Haiku 5.5 outperforms GPT-6 Luna across multiple benchmarks including computer use, knowledge work, frontier coding, and visual reasoning.72% similarUnverifiedIn 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.70% similarUnverifiedIn the GPT-5.6 family, Sol is the most powerful model targeting complex agentic workflows, Terra is a faster model targeting everyday workflows, and Luna is the fastest and most economical model targeting high-frequency, large-batch tasks.70% similarUnverifiedOpenAI 官方称 GPT-5.6 索拉纳执行同等复杂度编程任务时仅消耗 GPT-5.5 三分之一的 Token70% similarUnverifiedIn automated business process testing, GPT-6 Sol outperformed Claude Opus 5, with a per-task cost of approximately 9% of the latter.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/987475API
curl https://kongchang.com/api/v1/knowledge/claims/987475MCP
get_claim(id=987475)