Will there be evidence in 2025 that in April 2023, OpenAI had a GPT-4.5 or higher model?
Plus
20
Ṁ4252026
16%
chance
1D
1W
1M
ALL
Resolves YES if in 2025 there is a lot of evidence that OpenAI had a base model with a LM language-modelling loss lower than a model trained with 7X the compute using the GPT-4 scaling law.
This excludes retrieval and other "cheap to apply to GPT-4, known capability enhancements".
A "partially trained" model with a sufficiently low loss counts.
See https://twitter.com/taoroalin/status/1647049143519694848
This question is managed and resolved by Manifold.
Get
1,000and
3.00
1,0003.00Related questions
Related questions
Will OpenAI release GTP-4.5 before GPT-5?
73% chance
Will an open source model beat GPT-4 in 2024?
65% chance
Will there be a model that has a 75% win rate against the latest iteration of GPT-4 as of January 1st, 2025?
70% chance
Will OpenAI release a model better than GPT-4o without test time compute before EOY 2024?
74% chance
Will there be a OpenAI LLM known as GPT-4.5? by 2033
79% chance
Will we find out by 2026 that OpenAI abandoned an attempt to train GPT-5?
41% chance
Will OpenAI release GPT-5 before the end of 2024?
3% chance
Will OpenAI release a GPT-4 agent product before GPT-5?
67% chance
When will OpenAI release GPT-5?
Will OpenAI make GPT3.5 or 4 model weights open before 2025?
7% chance