Reasoning Mode News

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 DeepSeek's next reasoning model be called R3?

Feb 10, 8:33 AMSep 30, 11:59 AM

1.23%chance

3412190

OptionVotes

YES

8968

112

What is the main reason behind GPT-4o speed improvement relative to GPT-4 base model?

May 14, 12:17 PMJul 21, 9:59 PM

251944

OptionProbability

Smaller model size (hence, architecture/algorithm improvements)

More/better hardware allocated

Something related to low-level computation efficiency (for example, optimized frameworks)

Other

Better coarser-grained tokenizer

Will "large reasoning models" significantly outperform traditional LLMs in creative writing by 2026?

Jan 11, 7:11 AMJan 2, 7:59 AM

40.28%chance

141633

OptionVotes

YES

1218

821

Will DeepSeek's next reasoning model be open-sourced?

Feb 18, 7:49 PMJan 21, 7:59 AM

88.16%chance

151290

OptionVotes

YES

273

Will a Kimi reasoning model top LMArena by EOY2025?

Jul 17, 5:25 AMDec 31, 11:59 PM

25.59%chance

9731

OptionVotes

YES

1705

586

Before Feb 2026, will a transformer based reasoning model >1800 elo be able to explain 3+ chess lines at any position?

Jan 28, 10:50 PMFeb 2, 6:59 AM

55.1%chance

6485

OptionVotes

YES

1108

903

Will Hierarchical Reasoning Models scale?

Aug 3, 11:54 PMJan 1, 11:59 PM

48.73%chance

596

OptionVotes

YES

103

Will "Shutdown Resistance in Reasoning Models" make the top fifty posts in LessWrong's 2025 Annual Review?

Jul 7, 12:36 PMFeb 1, 12:00 AM

19%chance

372

OptionVotes

YES

1052

730

A top-three AI lab delays a frontier model release six months for safety reasons?

May 19, 12:23 PMJan 31, 12:20 PM

42%chance

537

OptionVotes

YES

118

Will "Do reasoning models use their scratchpad like..." make the top fifty posts in LessWrong's 2025 Annual Review?

Jul 28, 11:13 PMFeb 1, 12:00 AM

14.16%chance

224

OptionVotes

YES

1002

989

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 Mode News

Prediction markets for Reasoning Mode

Prediction markets for Reasoning Mode

Prediction markets for Reasoning Mode

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

Will DeepSeek's next reasoning model be called R3?

What is the main reason behind GPT-4o speed improvement relative to GPT-4 base model?

Will "large reasoning models" significantly outperform traditional LLMs in creative writing by 2026?

Will DeepSeek's next reasoning model be open-sourced?

Will a Kimi reasoning model top LMArena by EOY2025?

Before Feb 2026, will a transformer based reasoning model >1800 elo be able to explain 3+ chess lines at any position?

Will Hierarchical Reasoning Models scale?

Will "Shutdown Resistance in Reasoning Models" make the top fifty posts in LessWrong's 2025 Annual Review?

A top-three AI lab delays a frontier model release six months for safety reasons?

Will "Do reasoning models use their scratchpad like..." make the top fifty posts in LessWrong's 2025 Annual Review?

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

Latest stories