blackened / t-lite

t-lite Model of T-bank rus https://t.me/machinelearning_rus/18728

8B

489 Pulls Updated 2 months ago

2 months ago

f7dfa19f99d5 · 16GB

model
llama
·
8.03B
·
F16
template
{{ if .System }}<|start_header_id|>system<|end_header_id|> {{ .System }}<|eot_id|>{{ end }}{{ if .Prompt }}<|start_header_id|>user<|end_header_id|> {{ .Prompt }}<|eot_id|>{{ end }}<|start_header_id|>assistant<|end_header_id|> {{ .Response }}<|eot_id|>

Readme

Converted for Ollama by Evgeny Ryumin.

Specialists from the T-Bank Center for Artificial Intelligence have released a new language model in the 7-8 billion parameter range. According to internal and industry benchmarks, the model outperforms all existing counterparts in its class.

The T-lite model was introduced at the inaugural Turbo ML Conf, T-Bank’s first conference on machine learning. T-lite is a development tool that allows companies to create LLM applications for their own use without transferring data to third parties. These applications can include support assistants for handling requests, answering questions, etc. For example, an LLM can be used for voice-order processing.

image.png