Unverified50% confidenceBenchmarkExact time
With context management enabled, Iris-mini achieves 82.2%, 84.8%, 86.9%, and 52.3% on BrowseComp, BrowseComp-ZH, DeepSearchQA, and HLE respectively.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/12/2026
First Seen
Valid until: 12/11/2026
Sources
Related Entities
Related Claims
UnverifiedIMGNet achieved 76.85% vs. 74.62% on CPLFW and 78.80% vs. 77.22% on AgeDB-30, with a more pronounced advantage on higher-difficulty benchmarks65% similarUnverifiedOn the comprehensive benchmark, IMG Sign achieves 81.02% while cosine similarity achieves 79.49%; on LFW the figures are 96.27% and 95.53% respectively64% similarUnverifiedOn TruthfulQA, 8-bit scored 85.6%, slightly higher than OQ4E's 84.8%.63% similarUnverifiedAt small scale, the FLOPs share of the MLP layer is approximately 44%, rising to approximately 80% when scaled to 175B parameters61% similarUnverifiedDeepSeek-V3的技术报告显示其MTP机制在多种任务上的平均接受率约为85-90%58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/915232API
curl https://kongchang.com/api/v1/knowledge/claims/915232MCP
get_claim(id=915232)