Unverified50% confidenceFactExact time
When asked directly whether the total was visible, Claude, Gemini 3 Flash, and Flash-Lite answered across all 48 receipts that the total was not visible and could not be calculated, yet when asked directly for the total, they still fabricated one 21–44% of the time.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFabrication rates when the total is fully obscured: Claude Sonnet 5 at 38%, Gemini 3 Flash at 21%, Gemini 3.1 Flash-Lite at 44%, GPT-5.4 nano at 73%; fabrication rates when the total is half-obscured: Claude at 92%, the other three models all at 100%.66% similarUnverifiedQwen3.8 Flash Next has 125 billion total parameters, with only approximately 6 billion parameters activated per token during inference60% similarUnverifiedAgnes-3.0-Flash has a total of 72 layers, consisting of 54 delta-rule recurrent layers and 18 global attention layers arranged in a 3:1 alternating pattern, with a hidden dimension of 5120 and a vocabulary size of 248,32060% similarUnverifiedThe free version of Claude may hit its usage limit after just two or three prompts, whereas ordinary users of ChatGPT or Gemini often do not encounter limits.58% similarUnverified千问三235B MoE总参数量2350亿，但每次推理实际激活的参数约22B58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/997262API
curl https://kongchang.com/api/v1/knowledge/claims/997262MCP
get_claim(id=997262)