blossom-v5-14B: benchmark results
A conversational instruction fine-tune released by Azure99 based on Qwen1.5-14B, using Azure99/blossom-chat-v3, Azure99/blossom-math-v4, Azure99/blossom-wizard-v3, Azure99/blossom-orca-v3. Training uses two stages; dataset example counts and epochs are not token or compute disclosures. The card says its training data was distilled from gpt-4-0125-preview. Provider: Azure99. Access: Open.
Aggregate Rating 1423 ± 37, rank #5016 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 68.45 | Accuracy (%) (5-shot) | 91.5 |
| Open LLM Leaderboard v1 - GSM8K | 67.78 | Accuracy (%) (5-shot) | 84.4 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.89 | MC2 (%) (0-shot) | 63.8 |
| Open LLM Leaderboard v1 - HellaSwag | 80.72 | Normalized accuracy (%) (10-shot) | 42.1 |
| Open LLM Leaderboard v1 - WinoGrande | 75.14 | Accuracy (%) (5-shot) | 40.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 58.45 | Normalized accuracy (%) (25-shot) | 39.8 |
Interactive version: theaggregate.ai/model?slug=blossom-v5-14b · How It Works · Data refreshed daily, snapshot 2026-11-08.