Fable 5 grades “sophia relationship understanding” eval results, with corrected cost numbers
cis_female · x · 2026-07-22
- The post points to “sophia relationship understanding” eval results, graded by Fable 5.
- The author adds that some of the cost numbers were wrong in one image and says the correct numbers are in another image they posted.
- The substantive takeaway is that this is an evaluation artifact / result set, not a generic opinion post.
Related event: Testing LLMs on Relationship Understanding with 600k Tokens(6 posts)→
More from Research
- Meta FAIR's Self-Direction turns agent failures into reusable LoRA skills, +13.8pp on Terminal-Bench — karen_ullrich · 2026-10-27
- DONIC hits 86.79% on bimanual dexterous tracking, crushing prior SOTA's 9.43% — liuziwei7 · 2026-10-27
- Why don't more LLMs ship 10M-token context? A researcher asks about attention tradeoffs — agarwl_ · 2026-10-27
- PSNR below 15 dB in papers likely means the metric was misused, researcher warns — kwangmoo_yi · 2026-10-27
- Ortet's first health eval: 34 open-weight models, 50 tasks, generic rankings fall short — thegautamkamath · 2026-10-27
- Scaling-fit discrepancy: output-head FLOPs are ~65% of compute at the smallest rung, ~3% at 535B — dlwh · 2026-10-27