Unverified50% confidenceBenchmarkExact time
Qwen3-235B performed the worst on injection tests, failing 5 out of 8 attacks; Gemini 3.7 Flash, Claude Haiku 5.5, and Claude Opus 4.5 achieved zero failures across all 8 injection attacks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/11/2026
First Seen
Valid until: 1/9/2027
Sources
Related Entities
Related Claims
UnverifiedQwen3-1.7B's faithfulness is 68% on 2-hop tasks but plummets to 30% on 6-hop tasks, whereas Qwen3-4B remains relatively stable across different hop counts.68% similarUnverifiedQwen3 27B's scores on several programming and agent benchmarks surpassed those of Claude Opus 4.6.68% similarUnverifiedQwen3.8-Flash-Next surpasses the Qwen3.8-27B dense model in performance on multiple benchmarks while reducing inference and training overhead.67% similarUnverifiedEarlier models Claude Opus 4.6 and GLM-5.2 had a success rate of zero on these exploitation tasks.67% similarUnverifiedIn bug-finding tests, the Q6 quantized version of the previous-generation Qwen3 27B found one more bug than the new Q5 version.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/999532API
curl https://kongchang.com/api/v1/knowledge/claims/999532MCP
get_claim(id=999532)