FRIDA needs about 2.6 GB of memory at Q8_0 (a 0.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 6 quants, from 0.4 GB (TQ1_0) to 3.3 GB (F32).
FRIDA – a 0.823B parameter Russian‑English text embedding model
FRIDA is a full‑scale finetuned text embedding model based on the T5 encoder. It is trained on a Russian‑English dataset and supports embedding tasks with a maximum context of 512 tokens.
Good at
- Embedding Russian text
- Embedding English text (quality unknown)
- CLS pooling with prefixes
Limits
- 512 token context limit
- Designed for Russian, English quality uncertain
- No vision support
Best for
- Semantic search in Russian
- Paraphrase mining
- Document categorization
Who it’s for. Developers building search or recommendation systems that require Russian‑English embeddings. Researchers needing a lightweight embedding model for Russian text.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 0.4 GB | 1.0 GB | Fits | Get file → | |
| TQ2_0 | 0.5 GB | 1.0 GB | Fits | Get file → | |
| Q8_0 | 0.9 GB | 1.5 GB | Fits | Get file → | |
| BF16 | 1.7 GB | 2.2 GB | Fits | Get file → | |
| F16 | 1.7 GB | 2.2 GB | Fits | Get file → | |
| F32 | 3.3 GB | 3.9 GB | Fits | Get file → |
For 8 GB VRAM, use TQ1_0 (0.44 GB, 2.1 GB memory) or TQ2_0 (0.47 GB, 2.2 GB memory). For 16 GB VRAM, TQ1_0 or TQ2_0 remain suitable. Higher precision quants (Q8_0, BF16, F16, F32) require more memory and are not recommended for low‑budget GPUs.
What GPU do you need for FRIDA?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is BF16 (1.7 GB file, about 3.4 GB in use).
- 6 GB VRAM: the best quant that fits is F32 (3.3 GB file, about 5.0 GB in use).
Which FRIDA quant should you choose?
Q8_0 (0.9 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is TQ1_0 at 0.4 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about FRIDA
How much VRAM does FRIDA need?
About 2.6 GB at Q8_0 with an 8K context window. That is the 0.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run FRIDA on an 8 GB GPU?
Yes, with F32: the 3.3 GB file needs about 5.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of FRIDA GGUF?
It depends on the quant: TQ1_0 is 0.4 GB, TQ2_0 is 0.5 GB, Q8_0 is 0.9 GB, BF16 is 1.7 GB, F16 is 1.7 GB, F32 is 3.3 GB.
What context length does FRIDA support?
The GGUF header lists a trained context length of 512 tokens (512). The KV cache grows with the context you use: about 1.2 GB at 8K and 0.1 GB at 512.
Where can I download FRIDA GGUF?
The files are on Hugging Face in the evilfreelancer/FRIDA-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does FRIDA use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run FRIDA on 8 GB VRAM?
Yes, the TQ1_0 or TQ2_0 quantizations fit within 8 GB VRAM with an estimated 2.1–2.2 GB memory usage.
Does FRIDA support English text?
It is pre‑trained on Russian‑English data, but the quality in English is unknown.
What is the maximum input length?
The model accepts up to 512 tokens per input.