Expired70% confidenceFactTime unknown
The independent tester ran tests on Claude 4.5 Haiku more than five times repeatedly, and the results consistently showed poor performance.
1
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedClaude Haiku 4.5 failed all 5 programming task tests80% similarUnverifiedClaude Haiku 4.5 failed to achieve a chain explosion effect in all 6 tests of the firecracker chain explosion test79% similarUnverifiedIn an independent tester's overall rankings, Claude 4.5 Haiku placed 34th and 37th respectively, making it the lowest-scoring next-generation model among mainstream AI labs.71% similarUnverifiedClaude Haiku 4.5 has a lower false refusal rate than Sonnet 4.5 and Opus 4.1, making it one of the safest models Anthropic has released to date.66% similarUnverifiedIn actual performance tests, GLM 4.6 far outperformed Claude 4.5 Haiku.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24361API
curl https://kongchang.com/api/v1/knowledge/claims/24361MCP
get_claim(id=24361)