rubert-tiny2
This is an updated version of cointegrated/rubert-tiny: a small Russian BERT-based encoder with high-quality sentence embeddings. This post in Russian gives more details.
Params
30 M
Context
2,048
Downloads 30d
460 K
Likes
184
Download history
daily snapshots · 81 days
▲ 48 K in the last 30 days (11.6%)
460 K367 K
Oct 12Oct 22Nov 1Nov 10
460 K367 K
Aug 22Sep 18Oct 15Nov 10
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | f32 | 0.1 GB | 0.6 GB | ✅ Runs comfortably |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Specifications
- Architecture
- BertForPreTraining
- Parameters
- 30 M
- Tensor type
- F32
- Context length
- 2,048
- Vocabulary
- 83,828
- Layers / heads
- 3 / 12
- Licence
- mit
- First seen on the Hub
- 2022-03-02
- Training datasets
- undisclosed
- Added to our catalog
- 2026-08-22
Compare with any sentence-similarity model