UltraLM-13B-v2.0: benchmark results
The distinct UltraLM-13B-v2.0 release announced by the UltraChat project on September 26, 2023. Its sparse checkpoint card does not establish equivalence with the original UltraLM-13B delta release. Provider: Other. Released 2023-09-26. Access: Open.
Aggregate Rating 1361 ± 28, rank #7407 of 10916 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - ARC Challenge | 62.63 | Normalized accuracy (%) (25-shot) | 59 |
| AlpacaEval 1.0 | 86.28 | Win Rate (%) | 58.4 |
| Open LLM Leaderboard v1 - WinoGrande | 76.48 | Accuracy (%) (5-shot) | 48.5 |
| Open LLM Leaderboard v1 - HellaSwag | 81.49 | Normalized accuracy (%) (10-shot) | 45.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 49.48 | MC2 (%) (0-shot) | 44.9 |
| Open LLM Leaderboard v1 - GSM8K | 26.08 | Accuracy (%) (5-shot) | 44.4 |
| AlpacaEval | 63.78 | Length-Controlled Win Rate (%) | 43.9 |
| Open LLM Leaderboard v1 - MMLU | 56.17 | Accuracy (%) (5-shot) | 38.4 |
| AlpacaEval 2.0 | 9.13 | LC Win Rate (%) | 19.4 |
Interactive version: theaggregate.ai/model?slug=ultralm-13b-v2-0 · How It Works · Data refreshed daily, snapshot 2026-11-08.