ππͺπΎπ‘π‘πΎ
@gm8xx8
187 Following
130641 Followers
2 replies
2 recasts
11 reactions
1 reply
0 recast
4 reactions
1 reply
0 recast
5 reactions
1 reply
0 recast
3 reactions
2 replies
3 recasts
15 reactions
1 reply
6 recasts
10 reactions
2 replies
0 recast
6 reactions
1 reply
1 recast
7 reactions
OpenAIβs o1 update enhances reasoning through reinforcement learning, enabling step-by-step problem-solving similar to human thought. The longer it βthinks,β the better it performs, it introduces a new scaling paradigm beyond pretraining. Rather than relying solely on prompting, o1βs chain-of-thought reasoning improves with adaptive compute, which can be scaled at inference time.
- o1 outperforms GPT-4o in reasoning, ranking in the 89th percentile on Codeforces.
- It uses chain-of-thought to break down problems, correct errors, and adapt, though some specifics remain unclear.
- Excels in areas like data analysis, coding, and math.
- o1-preview and o1-mini models are available now, with evals proving itβs not just a one-off improvement. Trusted API users will have access soon.
- Results on AIME and GPQA are strong, with o1 showing significant improvement on complex prompts where GPT-4o struggles.
- The system card (https://openai.com/index/openai-o1-system-card/) showcases o1βs best capabilities. 5 replies
5 recasts
28 reactions
3 replies
2 recasts
7 reactions
1 reply
0 recast
9 reactions
1 reply
0 recast
6 reactions
2 replies
2 recasts
7 reactions
0 reply
0 recast
10 reactions
0 reply
0 recast
10 reactions
3 replies
1 recast
9 reactions
2 replies
0 recast
6 reactions
2 replies
2 recasts
17 reactions
3 replies
1 recast
13 reactions
2 replies
0 recast
7 reactions