badger-l3-instruct-32k: benchmark results
maldv Badger Llama 3 Instruct 32k is a recursive pairwise Fourier-interpolation merge. The publisher describes a fourfold RoPE configuration extension and reports coherent use at 16k to 32k context; this is an extension, not a claim of native pretraining context. Provider: maldv. Access: Open.
Aggregate Rating 1431 ± 34, rank #4636 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 72.4 | Accuracy (%) (5-shot) | 97.8 |
| Open LLM Leaderboard v1 - MMLU | 67.13 | Accuracy (%) (5-shot) | 88.9 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 55.02 | MC2 (%) (0-shot) | 64 |
| Open LLM Leaderboard v1 - ARC Challenge | 63.65 | Normalized accuracy (%) (25-shot) | 63.7 |
| Open LLM Leaderboard v1 - WinoGrande | 77.35 | Accuracy (%) (5-shot) | 55.2 |
| Open LLM Leaderboard v1 - HellaSwag | 81.4 | Normalized accuracy (%) (10-shot) | 44.9 |
Interactive version: theaggregate.ai/model?slug=badger-l3-instruct-32k · How It Works · Data refreshed daily, snapshot 2026-11-08.