Unverified50% confidenceBenchmarkExact time
Independent benchmarks show that LLM hallucination rates are roughly between 15% and over 50%, depending on the task and model.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedIn multiple benchmarks, even the most advanced models can have hallucination rates of 15%–30% in specialized technical domains.77% similarUnverifiedIn real physical laboratory validation of molecular design, the model achieved a hit rate of nearly 50% across 12 targets, compared to the industry average of approximately 10% to 15%.68% similarUnverifiedWhen testing LLM citation behavior, the same prompt should be run at least 20–50 times, ideally expanded to more than 100 times.68% similarUnverified2024 年的多项基准测试表明，经过严格安全对齐的模型在代码生成任务上的拒绝率比基座模型高出 15-30%，在创意写作任务上的自我审查率高达 40%68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1069296API
curl https://kongchang.com/api/v1/knowledge/claims/1069296MCP
get_claim(id=1069296)