GPQA News

Nvidia Unveils 9-Billion-Parameter Nemotron Nano Model With 6× Speed Boost

7 months ago

In what year will AI achieve a score of 95% or higher on the GPQA benchmark?

Jul 13, 10:10 PMJan 1, 11:59 PM

49521

OptionProbability

2026

2027

2028

2029

2025

Is scale unnecessary for intelligence (<10B param human-competitive STEM model before 2027)?

Dec 4, 4:35 AMJan 2, 4:59 AM

20.03%chance

217990

OptionVotes

YES

1998

500

What year will an o3-mini level model first run on a phone?

Jun 24, 8:26 PMJan 1, 11:59 PM

154860

OptionProbability

2026

2025

2027

2028

2029 or Later

Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027?

Jan 18, 12:11 AMJan 2, 7:59 AM

94.62%chance

304338

OptionVotes

YES

3951

504

By which date will the state-of-the-art LLM use latent space to reason?

Nov 9, 5:02 AMJan 2, 5:01 AM

163510

OptionProbability

Jan 1, 2029

Jan 1, 2028

Jan 1, 2027

Is scale unnecessary for intelligence (<10B param human-competitive STEM model before 2030)?

Nov 19, 4:05 AMJan 2, 4:59 AM

72.44%chance

262593

OptionVotes

YES

1621

617

Will the gap between open-weights and frontier models on GPQA Diamond be at most 7%?

Aug 5, 9:17 AMDec 31, 10:59 PM

5.72%chance

81095

OptionVotes

YES

406

Why is Claude 3.5 Sonnet such a good model for its size?

Jul 11, 8:22 AMJul 19, 6:59 AM

8547

OptionProbability

Pretraining data composition

Doesn't use any scale.ai training data

Offline policy learning RLHf

Task vectors like golden gate Claude

Will any Qwen 3.5 model beat Gemini 3 Pro's Artificial Analysis intelligence score by the end of March?

Feb 17, 12:52 PMMar 31, 10:59 PM

30.09%chance

389

OptionVotes

YES

152

Latest stories

Aug 18, 2025
Aug 18, 2025
AI Modeling
AI Products
AI
Nvidia Unveils 9-Billion-Parameter Nemotron Nano Model With 6× Speed Boost
Authors
12
7 months ago

Join our channels

Get the latest stories live on any device.

WhatsApp
Top Stories
Join
Telegram
Top Stories
Join

GPQA News

Top stories

Prediction markets for GPQA

Prediction markets for GPQA

Prediction markets for GPQA

In what year will AI achieve a score of 95% or higher on the GPQA benchmark?

Is scale unnecessary for intelligence (<10B param human-competitive STEM model before 2027)?

What year will an o3-mini level model first run on a phone?

Will AIs beat human experts in question-answering on the GPQA benchmark before January 1st, 2027?

By which date will the state-of-the-art LLM use latent space to reason?

Is scale unnecessary for intelligence (<10B param human-competitive STEM model before 2030)?

Will the gap between open-weights and frontier models on GPQA Diamond be at most 7%?

Why is Claude 3.5 Sonnet such a good model for its size?

Will any Qwen 3.5 model beat Gemini 3 Pro's Artificial Analysis intelligence score by the end of March?

Latest stories

Join our channels

WhatsApp

Telegram