Reasoning Tokens News

Will an AI score over 80% on FrontierMath Benchmark in 2025

Nov 9, 1:06 AMDec 31, 10:59 PM

10%chance

113116754

OptionVotes

YES

1667

914

Will the state-of-the-art AI model use latent space to reason by 2026?

Dec 14, 1:06 AMJan 2, 1:00 AM

15.31%chance

2629485

OptionVotes

YES

23517

4252

Will GPT-5 be able to solve A::B system puzzles consistently

Apr 5, 2:04 PMJan 1, 9:59 PM

15.46%chance

5711934

OptionVotes

YES

2194

231

Will scaling current methods be enough to eliminate LLM hallucination?

Feb 27, 11:31 PMJan 2, 5:59 AM

15.34%chance

533171

OptionVotes

YES

1805

749

Will JEPA models become mainstream by the end of 2025?

Aug 24, 4:55 AMJan 1, 4:59 AM

9.71%chance

213084

OptionVotes

YES

2154

787

Before 2028, will any AI model achieve the same or greater benchmarks as o3 high with <= 1 million tokens per question?

Dec 21, 6:17 PMJan 2, 4:59 AM

88.14%chance

102611

OptionVotes

YES

2726

367

Will top open-weight LLMs in 2025 reason opaquely?

Dec 11, 6:09 AMJan 2, 7:59 AM

24.16%chance

271808

OptionVotes

YES

1772

564

When will the next paradigm in LLMs (after reasoning) be released?

Feb 17, 4:24 AMJan 1, 7:59 AM

31122

OptionProbability

Q1/Q2 '26

Q3/Q4 '25

Q3/Q4 '26

Q1/Q2 '27

Q1/Q2 '25

Q3/Q4 '27

2028 or later

Before 2026, What will be true of OpenAI's Claimed IMO Gold Performance?

Jul 21, 2:15 AMDec 31, 11:59 PM

9267

OptionProbability

At least one former IMO medalist will review the model's answers and claim it did not actually achieve Gold

The model that achieved it was trained with a new reinforcement learning algorithm

The model that achieved it could earn at least bronze using no more than 100,000 reasoning tokens per question

It was achieved with the same model OpenAI used to get second place in AtCoder

The breakthrough is mostly the result of superior test-time scaling methods

I will consider the techniques used to achieve it at least as big of a breakthrough as strawberry

It was achieved by a model that does not use a standard transformer architecture

Will OpenAI release a model which generates images using reasoning / inference-time scaling before 2026?

Dec 21, 9:28 PMJan 3, 10:59 PM

48%chance

7207

OptionVotes

YES

104

Will the OpenRouter LLM input:output token ratio exceed 100 in June 2026?

Jul 22, 2:10 PMJun 7, 11:59 PM

34.79%chance

265

OptionVotes

YES

137

Non-token based CoTs in SOTA reasoning models EOY 2026?

Jul 21, 5:03 PMDec 31, 11:59 PM

115

OptionProbability

All: All SOTA reasoning models use non-token-based CoT

Some: At least one, but not all, SOTA reasoning models use non-token- based CoT

None: No SOTA reasoning models use non-token-based CoT

Reasoning Tokens News

Prediction markets for Reasoning Tokens

Prediction markets for Reasoning Tokens

Prediction markets for Reasoning Tokens

Will an AI score over 80% on FrontierMath Benchmark in 2025

Will the state-of-the-art AI model use latent space to reason by 2026?

Will GPT-5 be able to solve A::B system puzzles consistently

Will scaling current methods be enough to eliminate LLM hallucination?

Will JEPA models become mainstream by the end of 2025?

Before 2028, will any AI model achieve the same or greater benchmarks as o3 high with <= 1 million tokens per question?

Will top open-weight LLMs in 2025 reason opaquely?

When will the next paradigm in LLMs (after reasoning) be released?

Before 2026, What will be true of OpenAI's Claimed IMO Gold Performance?

Will OpenAI release a model which generates images using reasoning / inference-time scaling before 2026?

Will the OpenRouter LLM input:output token ratio exceed 100 in June 2026?

Non-token based CoTs in SOTA reasoning models EOY 2026?

Latest stories