Unverified50% confidenceBenchmarkExact time
With the open-source model GPT 4.6 V4 Flash paired with H1, it achieves a 43% solve rate on the X line, exceeding the 34% of the specially optimized LeanStraw 1.5.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/17/2026
First Seen
Valid until: 12/1/2026
Sources
Simple Loop Tops Automated Theorem Proving: The Bitter Lesson Revisited
bilibili熊二等兵11/1/2026
Related Entities
Related Claims
VerifiedV4.1-Flash scores 1 point higher than a GPT-5.6-level model in comparison.80% similarUnverifiedOn frontier models such as GPT-5.4, Gemini-3.5-Flash, and Qwen3.6-27B, the page-image-based approach comprehensively outperforms OCR-based baselines and requires fewer tool calls.73% similarUnverifiedOpenAI's GPT-4o mini / GPT-4o and Google's Flash / Pro series likewise make layered trade-offs among latency, cost, and capability.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1180795API
curl https://kongchang.com/api/v1/knowledge/claims/1180795MCP
get_claim(id=1180795)