BTZSC - Sentiment: leaderboard
BTZSC sentiment slice measures macro-F1 on zero-shot sentiment classification datasets.
Metric: Macro-F1 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 8B | 89.87 | #674 |
| 2 | Qwen3-Embedding-8B | 89.16 | #685 |
| 3 | Qwen 3 4B | 88.32 | #829 |
| 4 | Mistral Nemo Instruct (2407) | 84.13 | #936 |
| 5 | Qwen3-Embedding-0.6B | 81.03 | #802 |
| 6 | Gemma 3 1B (IT) | 51.94 | #1504 |
| 7 | Phi-4 Mini Instruct | 49.01 | #1245 |
| 8 | Llama 3.2 3B Instruct | 45.58 | #1352 |
| 9 | Gemma 3 270M (IT) | 42.08 | #1581 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=btzsc-sentiment · How It Works · Data refreshed daily, snapshot 2026-10-20.