LegendaryFred GGUF

The legendary FredT5 XL. The BEST nick analyzer ever.

End of 2025 year: still NO models (even 100B+) can beat FRED-T5-1.7b (2023 SOTA for Russian language) in sense of trash nicknames analyzing...

Example outputs

Хабрастатьи тут и немного тут.

Потом, если получится хоть 1 рабочий конверт, засуну результаты самого gguf'а

Getting started

You can use LM Studio or llama.cpp for example. In LM Studio search for model and download the Q4 version, then write some dialog, LLM will continue better if dialog is not too tiny (need 50-80+ words).

Origin

This is Llama.cpp compatible versions of an original 1.7B model + some tokenizer patches, prompt templates.

  • Base model
  • Finetune of Base model I used for this convertation

Developing

Converted with PAIN using patched llama cpp

Downloads last month
21
GGUF
Model size
2B params
Architecture
t5
Hardware compatibility
Log In to add your hardware

5-bit

16-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for 3ndetz/fred-t5-xl-remastered-GGUF

Quantized
(1)
this model