/Tech1h ago

Fable Shows Strong Big Model Smell Over GPT-4.5 in Pre-Training Debate

311102.7K

#1215

Original post

Lisan al Gaib@scaling01#1215inTech

but pre-training token amount and quality matter of course

so in that sense Fable should have much more big model smell than GPT-4.5

Lisan al Gaib@scaling01

if by big model smell you literally just mean total params then the ranking should probably look something like this:

GPT-4.5 Claude Fable Gemini 3/3.1 Pro Opus 4 / 4.1 Grok 3 / 4 Opus 3 GPT-5.5 Opus 4.5 - 4.8 GPT-4 DeepSeek-V4 Pro Kimi-K2

1:46 PM · Jun 14, 2026 · 1.2K Views

Sentiment

Users reacted with disgust to claims that pre-training scale gives Claude Fable a stronger big model smell than GPT-4.5, calling the model smelly, stinky, and reeking.

Pos

0.0%

Neg

100.0%

1 comments with sentiment.

Cluster Engagement

Posts from X

Most Activity

VIEWS1.5KLIKES6REPLIES1

Lisan al Gaib@scaling01

Fable is really smelly, stinky actually reaking

Lisan al Gaib@scaling01

but pre-training token amount and quality matter of course

so in that sense Fable should have much more big model smell than GPT-4.5

1h1.5K60

Lisan al Gaib@scaling01

reeking? idk

1h8632