blossom-v1-3B: benchmark results
A conversational instruction fine-tune released by Azure99 based on Bloom-3b, using Azure99/blossom-chat-v1. Its card describes 60K Blossom chat examples. Provider: Azure99. Access: Open.
Aggregate Rating 1152 ± 36, rank #9934 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 43.45 | MC2 (%) (0-shot) | 23.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 36.86 | Normalized accuracy (%) (25-shot) | 14.2 |
| Open LLM Leaderboard v1 - MMLU | 26.7 | Accuracy (%) (5-shot) | 13.4 |
| Open LLM Leaderboard v1 - HellaSwag | 55.1 | Normalized accuracy (%) (10-shot) | 12.2 |
| Open LLM Leaderboard v1 - WinoGrande | 58.88 | Accuracy (%) (5-shot) | 12 |
| Open LLM Leaderboard v1 - GSM8K | 0.38 | Accuracy (%) (5-shot) | 11.9 |
Interactive version: theaggregate.ai/model?slug=blossom-v1-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.