Reasoning benchmark
LAMBADA leaderboard
As of October 2026, Falcon-180B has the highest published LAMBADA score on Noometry at 79.8%, out of 9 models with results.
Last verified
About LAMBADA
Predicting the last word of a passage, which needs the broader context.
- Category
- Reasoning
- Introduced
- 2016
- Format
- Word prediction
- Unit
- Percent (random guessing ≈ 0%)
- Official site
- zenodo.org
Top 9 models
- Falcon-180B 79.8%
- Llama 2-70B 78.9%
- Falcon-40B 77.3%
- Llama 2-13B 76.5%
- Llama 13b 75.2%
- Falcon-7B 74.9%
- Llama 2-7B 73.3%
- Qwen-14B 71.1%
- Qwen-7B 67.9%
Sponsored placements are available on pages like this one. Advertise on Noometry
All results
| # | Model | Provider | Score | Setting | Source | Date |
|---|---|---|---|---|---|---|
| 1 | Falcon-180B |  Technology Innovation Institute | 79.8% | Epoch AI | ||
| 2 | Llama 2-70B |  Meta | 78.9% | Epoch AI | ||
| 3 | Falcon-40B | Technology Innovation Institute | 77.3% | Epoch AI | ||
| 4 | Llama 2-13B | Meta | 76.5% | Epoch AI | ||
| 5 | Llama 13b | Meta | 75.2% | Epoch AI | ||
| 6 | Falcon-7B | Technology Innovation Institute | 74.9% | Epoch AI | ||
| 7 | Llama 2-7B | Meta | 73.3% | Epoch AI | ||
| 8 | Qwen-14B |  Alibaba (Qwen) | 71.1% | Epoch AI | ||
| 9 | Qwen-7B | Alibaba (Qwen) | 67.9% | Epoch AI |
Compare the leaders
Frequently asked questions
What does LAMBADA measure?
Predicting the last word of a passage, which needs the broader context.
Which model has the highest LAMBADA score?
As of October 2026, Falcon-180B has the highest published LAMBADA score on Noometry at 79.8%, out of 9 models with results.
What is the best open-weight model on LAMBADA?
Falcon-180B has the highest LAMBADA accuracy among open-weight models at 79.8%, ranking 1 of 9 overall.