Expired50% confidenceEventTime unknown
A Bilibili content creator conducted systematic visual programming task tests on Claude Haiku 4.5
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Claude Haiku 4.5 Real-World Test: Failed All 5 Programming Tasks
bilibilikarminski-牙医
Related Claims
UnverifiedThe tester recommends using Claude Sonnet 3.5 or higher-tier models, rather than Claude Haiku 4.5, for scenarios involving code generation, logical reasoning, and spatial understanding68% similarUnverifiedAnthropic has publicly released a detailed System Card for Claude Haiku 4.5, covering training data sources, prompt engineering strategies, and benchmark results.68% similarUnverifiedIn the Pillow library RGB-to-HSV precision bug test, Claude Code using the Sonnet 4.6 model passed on the first attempt with clear logic, earning a perfect score.65% similarUnverifiedA Bilibili content creator named Jiuling conducted a head-to-head comparison test of five AI models for game development.65% similarUnverifiedA Bilibili content creator conducted hands-on testing of the DS4 project, covering frontend mini-games and STM32 embedded development scenarios.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/54457API
curl https://kongchang.com/api/v1/knowledge/claims/54457MCP
get_claim(id=54457)