falcon-rw-1B-chat: benchmark results
ericzzz’s conversational refinement of Falcon-RW-1B-Instruct-OpenOrca, trained on selected UltraChat and OpenChat ShareGPT data. It is a separate fine-tune from both the Falcon-RW base and the OpenOrca instruction model. Provider: ericzzz. Access: Open.
Aggregate Rating 1150 ± 36, rank #9963 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 1.67 | Accuracy (%) (5-shot) | 17.9 |
| Open LLM Leaderboard v1 - HellaSwag | 61.12 | Normalized accuracy (%) (10-shot) | 15.1 |
| Open LLM Leaderboard v1 - WinoGrande | 61.72 | Accuracy (%) (5-shot) | 14.9 |
| Open LLM Leaderboard v1 - ARC Challenge | 35.58 | Normalized accuracy (%) (25-shot) | 13.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 39.62 | MC2 (%) (0-shot) | 11.6 |
| Open LLM Leaderboard v1 - MMLU | 24.51 | Accuracy (%) (5-shot) | 4 |
Interactive version: theaggregate.ai/model?slug=falcon-rw-1b-chat · How It Works · Data refreshed daily, snapshot 2026-11-08.