Flan-PaLM 540B (Google) and Wu Dao 2.0 (Beijing Academy of Artificial Intelligence / BAAI) are both frontier AI models. Flan-PaLM 540B was published in October 2022 and Wu Dao 2.0 in May 2021.
Flan-PaLM 540B was trained on 2.5 x 10^24 FLOP, about 1.6x 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 | Flan-PaLM 540B | Wu Dao 2.0 |
|---|---|---|
| Organization | Beijing Academy of Artificial Intelligence / BAAI | |
| Published | Oct 20, 2022 | May 31, 2021 |
| Training compute | 2.5 x 10^24 FLOP | 1.5 x 10^24 FLOP |
| Parameters | 540B | 1.8T |
| Dataset size | 1.4B | 4.9T |
| Training hardware | Google TPU v4 | — |
| Chips used | 512 | — |
| Training time | 37 h | — |
| Training cost (2023 USD) | — | $3M |
| Training power draw | 348.3 kW | — |
| Accessibility | Unreleased | API access |
| Open weights | No | No |
| Country | United States | China |
