Early tester finds astra less creative and thorough than fable 5.1 on coding and analysis
felpix_ · x · 2026-09-06
User felpix shares first impressions of astra: the leaps others describe mostly come from visual or computer-use tasks, which he doesn't need. On coding, data analysis, and finance tasks, he found it less creative and less thorough than fable 5.1.
More from Models
- Matt Shumer: I 4x'd my Pro/Max subs as agentic models remove the attention bottleneck — mattshumer_ · 2026-09-07
- AI startup Felda dissects writing quality down to sentence rhythm and punctuation in its new benchmark — almmaasoglu · 2026-09-07
- Teaching Qwen Next 3D sculpting in Blender by having GPT Astra coach it via MCP — LegacyRemaster · 2026-09-07
- GPT-6 Astra beats Portal for ~$167 — the impressive part is 3D navigation, not puzzle solving — burny_tech · 2026-09-07
- omarsar0: the most open, easily tunable models will win agent adoption — omarsar0 · 2026-09-07
- omarsar0: personalized general intelligence will be built largely on open-source models — omarsar0 · 2026-09-07