smol_llama-220M-open_instruct: benchmark results
BEE-spoke-data release: An instruction fine-tune of smol_llama-220M-GQA using VMware/open-instruct, distinct from the OpenHermes fine-tune. Provider: BEE-spoke-data. Access: Open.
Aggregate Rating 1104 ± 43, rank #10681 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 44.06 | MC2 (%) (0-shot) | 25.8 |
| Open LLM Leaderboard v1 - MMLU | 26.11 | Accuracy (%) (5-shot) | 11.3 |
| Open LLM Leaderboard v1 - HellaSwag | 29.71 | Normalized accuracy (%) (10-shot) | 5.3 |
| Open LLM Leaderboard v1 - GSM8K | 0 | Accuracy (%) (5-shot) | 4.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 25 | Normalized accuracy (%) (25-shot) | 4.3 |
| Open LLM Leaderboard v1 - WinoGrande | 50.28 | Accuracy (%) (5-shot) | 3.2 |
Interactive version: theaggregate.ai/model?slug=smol-llama-220m-open-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.