GPT-4 (Mar 2023) (OpenAI) and Wu Dao 2.0 (Beijing Academy of Artificial Intelligence / BAAI) are both frontier AI models. GPT-4 (Mar 2023) was published in March 2023 and Wu Dao 2.0 in May 2021.
GPT-4 (Mar 2023) was trained on 2.1 x 10^25 FLOP, about 13.6x the compute of Wu Dao 2.0 at 1.5 x 10^24 FLOP. Training compute is the closest available proxy for how much was invested in a model, though it says nothing on its own about how well that compute was spent.
These two models share no benchmark on which both have been scored, so no direct performance comparison is possible here. The specification table below is a comparison of inputs, not of results.
| Field | GPT-4 (Mar 2023) | Wu Dao 2.0 |
|---|---|---|
| Organization | OpenAI | Beijing Academy of Artificial Intelligence / BAAI |
| Published | Mar 15, 2023 | May 31, 2021 |
| Training compute | 2.1 x 10^25 FLOP | 1.5 x 10^24 FLOP |
| Parameters | 1.8T | 1.8T |
| Dataset size | 5.4T | 4.9T |
| Training hardware | NVIDIA A100 SXM4 40 GB | — |
| Chips used | 25,000 | — |
| Training time | 2.3K h | — |
| Training cost (2023 USD) | $37M | $3M |
| Training power draw | 19.9 MW | — |
| Accessibility | API access | API access |
| Open weights | No | No |
| Country | United States | China |
