BTZSC - Embedding: leaderboard
BTZSC embedding-model slice measures zero-shot text classification macro-F1 for embedding methods.
Metric: Macro-F1 (%). Source: huggingface.co. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gte-large-en-v1.5 | 61.74 | |
| 2 | e5-large-v2 | 59.74 | |
| 3 | e5-base-v2 | 59.66 | |
| 4 | Qwen3-Embedding-8B | 59.13 | #685 |
| 5 | gte-modernbert-base | 58.64 | |
| 6 | gte-base-en-v1.5 | 58.42 | |
| 7 | Qwen3-Embedding-0.6B | 57.97 | #802 |
| 8 | e5-mistral-7b-instruct | 57.56 | |
| 9 | bge-base-en-v1.5 | 56.83 | |
| 10 | bge-large-en-v1.5 | 55.48 |
Interactive version: theaggregate.ai/benchmark?slug=btzsc-embedding · How It Works · Data refreshed daily, snapshot 2026-10-20.