You can deploy this model and start earning money today!

abrosimov-a-a/
YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF

Name: YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF
Availability: InStock

quantized

yandex/YandexGPT-5-Lite-8B-instruct

Views

No views yet

abrosimov-a-a/YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF

This model was converted to GGUF format from yandex/YandexGPT-5-Lite-8B-instruct using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

НЕ КАЧАЙТЕ ЭТУ МОДЕЛЬ И НЕ ПЫТАЙТЕСЬ "ВЛОБ" КВАНТОВАТЬ YandexGPT-5-Lite-8B-instruct!!!

LLAMA.CPP МЕНЯЕТ ШАБЛОН МОДЕЛИ, ИЗ-ЗА ЧЕГО ОНА НАЧИНАЕТ ВЕСТИ СЕБЯ НЕАДЕКВАТНО.

ОСОБЕННО, ЭТО ПРОЯВЛЯЕТСЯ ПРИ ИСПОЛЬЗОВАНИИ ИНСТРУМЕНТОВ (TOOLS), МОДЕЛЬ ДЕЙСТВИТЕЛЬНО "ВИДИТ ФИГУ"

У Yandex есть хорошая квантизованная модель

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

bash

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo abrosimov-a-a/YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF --hf-file yandexgpt-5-lite-8b-instruct-q6_k.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo abrosimov-a-a/YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF --hf-file yandexgpt-5-lite-8b-instruct-q6_k.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo abrosimov-a-a/YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF --hf-file yandexgpt-5-lite-8b-instruct-q6_k.gguf -p "The meaning to life and the universe is"

./llama-server --hf-repo abrosimov-a-a/YandexGPT-5-Lite-8B-instruct-Q6_K-GGUF --hf-file yandexgpt-5-lite-8b-instruct-q6_k.gguf -c 2048