Unverified50% confidenceBenchmarkExact time
During a demo walk of a long-sequence 3D benchmark test, the system's recognition accuracy for nearby targets was 2/2 and for distant targets was 3/10.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/7/2026
First Seen
Valid until: 10/21/2026
Sources
Related Entities
Related Claims
UnverifiedThe model gave the correct answer of 34 in a multimodal image recognition test counting pencils, at a speed of 55 tokens per second at that time.66% similarUnverifiedThe machine precision of double-precision floating-point numbers is approximately 2.22 × 10^-16.65% similarUnverified4Bit量化下模型的基准测试得分通常能保持原始精度的90%-95%64% similarUnverifiedThe authors tested degree-3 and degree-4 BF16 polynomial versions to balance accuracy and speed.63% similarUnverifiedPrecision Mode achieved an extraction accuracy rate of 94.7% on a benchmark covering approximately 9,000 complex documents.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983658API
curl https://kongchang.com/api/v1/knowledge/claims/983658MCP
get_claim(id=983658)