Falcon-180B (Technology Innovation Institute) and PaLM 2 (Google) are both frontier AI models. Falcon-180B was published in September 2023 and PaLM 2 in May 2023.
PaLM 2 was trained on 7.3 x 10^24 FLOP, about 2.0x the compute of Falcon-180B at 3.8 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 | Falcon-180B | PaLM 2 |
|---|---|---|
| Organization | Technology Innovation Institute | |
| Published | Sep 6, 2023 | May 10, 2023 |
| Training compute | 3.8 x 10^24 FLOP | 7.3 x 10^24 FLOP |
| Parameters | 180B | 340B |
| Dataset size | 3.5T | 3.6T |
| Training hardware | NVIDIA A100 SXM4 40 GB | Google TPU v4 |
| Chips used | 4,096 | — |
| Training time | 4.3K h | — |
| Training cost (2023 USD) | $11M | $5M |
| Training power draw | 3.3 MW | — |
| Accessibility | Open weights (restricted use) | API access |
| Open weights | Yes | No |
| Country | United Arab Emirates | United States |
