Will transformers still be the dominant DL architecture in 2026?

30

Ṁ1kṀ5.8k

resolved Jan 2

Resolved

YES

1H

6H

1D

1W

1M

ALL

Resolves true if I judge, based on the common opinion among deep learning researchers, that transformers remain the most popular architecture in deep learning research at the start of 2026. If the answer is not clear, resolves true if at least 50% of arXiv papers from 2025 on A.I. mention transformers, otherwise resolves false.

Market context

Get

1,000

to start trading!

🏅 Top traders

#	Trader	Total profit
1		Ṁ314
2		Ṁ243
3		Ṁ109
4		Ṁ94
5		Ṁ67

Sort by:

sold Ṁ10 YES

It seems plausible that OpenAI, Google, Anthropic will have moved away from transformers for their closed models by then, but we might not know about it.

@MalachiteEagle in that case it would not be "common opinion."

Apparently there is already a very similar market!

https://manifold.markets/LeoGao/will-transformer-based-architecture

People are also trading

Will transformer architectures lose their dominant position in deep learning before 2028?

Will superposition in transformers be mostly solved by 2026?

On January 1, 2027, a Transformer-like model will continue to hold the state-of-the-art position in most benchmark

Will the most capable, public multimodal model at the end of 2027 in my judgement use a transformer-like architecture?

Will Transformer-Based LLMs Make Up ≥75% of Parameters in the Top General AI by 2030?

Are Transformers “the last technology to do all of ML” (before AGI)?

Will there be over 10,000 Optimus robots working at Tesla before 2027?

[Will Brown] Will Transformative AI look like an abundance of specialized models, in August 2026?

Will the transformer architecture be replaced in SOTA LLMs by 2028?

When will a non-Transformer model become the top open source LLM?

Related questions

Will transformer architectures lose their dominant position in deep learning before 2028?

Will superposition in transformers be mostly solved by 2026?

On January 1, 2027, a Transformer-like model will continue to hold the state-of-the-art position in most benchmark

Will the most capable, public multimodal model at the end of 2027 in my judgement use a transformer-like architecture?

Will Transformer-Based LLMs Make Up ≥75% of Parameters in the Top General AI by 2030?

Are Transformers “the last technology to do all of ML” (before AGI)?

Will there be over 10,000 Optimus robots working at Tesla before 2027?

[Will Brown] Will Transformative AI look like an abundance of specialized models, in August 2026?

Will the transformer architecture be replaced in SOTA LLMs by 2028?

When will a non-Transformer model become the top open source LLM?

© Manifold Markets, Inc.•Terms•Privacy