WizardLM-13B-V1.0: benchmark results
The WizardLM team's original WizardLM-13B V1.0, distributed as diff weights to apply to a base LLaMA checkpoint, part of the WizardLM instruction-following model family. Provider: Microsoft. Released 2023-05-13. Access: Open.
Aggregate Rating 1294 ± 31, rank #1465 of 1635 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 50.55 | MC2 (%) (0-shot) | 48.7 |
| Open LLM Leaderboard v1 - HellaSwag | 80.88 | Normalized accuracy (%) (10-shot) | 42.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 57.25 | Normalized accuracy (%) (25-shot) | 37 |
| Open LLM Leaderboard v1 - GSM8K | 14.1 | Accuracy (%) (5-shot) | 36 |
| Open LLM Leaderboard v1 - WinoGrande | 74.11 | Accuracy (%) (5-shot) | 35 |
| Open LLM Leaderboard v1 - MMLU | 52.92 | Accuracy (%) (5-shot) | 32.2 |
| Open LLM Leaderboard - MuSR | 34.97 | Score | 12.3 |
| Open LLM Leaderboard - GPQA | 25.92 | Score | 11.7 |
| Open LLM Leaderboard - IFEval | 18.5 | Score | 9.9 |
| Open LLM Leaderboard - MMLU-Pro | 11.66 | Score | 7.9 |
| Open LLM Leaderboard - BBH | 29.13 | Score | 2.2 |
| Open LLM Leaderboard - MATH Level 5 | 0 | Score | 1.3 |
Interactive version: theaggregate.ai/model?slug=wizardlm-13b-v1-0 · How It Works · Data refreshed daily, snapshot 2026-11-05.