blossom-v3.1-yi-34B: benchmark results
A conversational instruction fine-tune released by Azure99 based on Yi-34B, using Azure99/blossom-chat-v1, Azure99/blossom-math-v2, Azure99/blossom-wizard-v1, Azure99/blossom-orca-v1. Training uses two stages; dataset example counts and epochs are not token or compute disclosures. Provider: Azure99. Access: Open.
Aggregate Rating 1468 ± 36, rank #2902 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 74.37 | Accuracy (%) (5-shot) | 96 |
| Open LLM Leaderboard v1 - WinoGrande | 82.08 | Accuracy (%) (5-shot) | 82.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 65.36 | Normalized accuracy (%) (25-shot) | 69.8 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 56.06 | MC2 (%) (0-shot) | 66.9 |
| Open LLM Leaderboard v1 - HellaSwag | 84.24 | Normalized accuracy (%) (10-shot) | 64.8 |
| Open LLM Leaderboard v1 - GSM8K | 46.85 | Accuracy (%) (5-shot) | 60.3 |
Interactive version: theaggregate.ai/model?slug=blossom-v3-1-yi-34b · How It Works · Data refreshed daily, snapshot 2026-11-08.