AI
12.24'24
Platform
-
Hugging Face: https://huggingface.co/
- quicktour: docs/transformers/quicktour
-
modal: https://modal.com/
Website
-
Chatbot Arena:
-
poe: https://poe.com/
-
deepseek: https://chat.deepseek.com/
-
Perplexity: https://labs.perplexity.ai/
Project
-
ollama: https://github.com/ollama/ollama
-
FastChat: https://github.com/lm-sys/FastChat
-
llama: https://ai.meta.com/llama/
-
llama Github: https://github.com/facebookresearch/llama
-
llama.cpp Port of Facebook's LLaMA model in C/C++
-
llama-gpt
-
whisper.cpp Port of OpenAI's Whisper model in C/C++
-
Personalizing LLM Responses
llama.cpp: Run on macOS
# use gpu
./main -m ~/Downloads/llama-2-7b-chat.Q4_K_M.gguf --prompt "what is quantum"
# disable Metal gpu
./main -ngl 0 -m ~/Downloads/llama-2-7b-chat.Q4_K_M.gguf --prompt "what is quantum"
Models
Models are usually downloaded from Hugging Face
Leaderboard
Serving
- vLLM: https://github.com/vllm-project/vllm
- webui: https://github.com/oobabooga/text-generation-webui
Check later
📖