Skeptic doubts Anthropic achieved large-factor bpb gains at fixed compute
teortaxesTex · x · 2026-11-14
Commenting on a circulating (admittedly vague) chart, the author argues:
- He doesn't believe Anthropic has improved algorithms enough to deliver a large-factor bpb improvement at isodata/isoflop conditions; if that were literally the case it would be an astonishing update.
- Other labs (e.g. TM, and even MAI's recent reports) have also claimed OOM+ bpb improvements at given scale, so this kind of algorithmic progress isn't monopolized by Anthropic even if exponents differ.
More from Models
- Together AI webinar dissects Ripple-65B coding model: dense vs MoE, ablations, RL runs — togethercompute · 2026-11-14
- Developer seeks free model alternatives after losing free DeepSeek access in Pi editor — nirurin · 2026-11-14
- Surge AI benchmark: every Claude model, even Haiku, beats every GPT on staff-level coding — echen · 2026-11-14
- AI cracks Gödel's 1,009-page notebooks, uncovering hundreds of unknown theorems in days — stevenstrogatz · 2026-11-14
- NVIDIA lands nvfp4 in large-scale pretraining under codename "yolo bundle" — SonglinYang4 · 2026-11-14
- Delip Rao mocks vendors testing safety in production, then rolling back updates — deliprao · 2026-11-14