lguzzon-scratchbook / ask-llm

Interact with an LLM service

Ask LLM

This is a straightforward, zero-dependency CLI tool to interact with any LLM service.

It is available in several flavors:

• Python version. Compatible with CPython or PyPy, v3.10 or higher.
• JavaScript version. Compatible with Node.js (>= v18) or Bun (>= v1.0).
• Clojure version. Compatible with Babashka (>= 1.3).
• Go version. Compatible with Go, v1.19 or higher.

Ask LLM is compatible with either a cloud-based (managed) LLM service (e.g. OpenAI GPT model, Groq, OpenRouter, etc) or with a locally hosted LLM server (e.g. llama.cpp, LocalAI, Ollama, etc). Please continue reading for detailed instructions.

Interact with the LLM with:

./ask-llm.py         # for Python user
./ask-llm.js         # for Node.js user
./ask-llm.clj        # for Clojure user
go run ask-llm.go    # for Go user

or pipe the question directly to get an immediate answer:

echo "Why is the sky blue?" | ./ask-llm.py

or request the LLM to perform a certain task:

echo "Translate into German: thank you" | ./ask-llm.py

Using Local LLM Servers

Supported local LLM servers include llama.cpp, Jan, Ollama, and LocalAI.

To utilize llama.cpp locally with its inference engine, ensure to load a quantized model such as Phi-3 Mini, LLama-3 8B, or OpenHermes 2.5. Adjust the environment variable LLM_API_BASE_URL accordingly:

/path/to/llama.cpp/server -m Phi-3-mini-4k-instruct-q4.gguf
export LLM_API_BASE_URL=http://127.0.0.1:8080/v1

To use Jan with its local API server, refer to its documentation and load a model like Phi-3 Mini, LLama-3 8B, or OpenHermes 2.5 and set the environment variable LLM_API_BASE_URL:

export LLM_API_BASE_URL=http://127.0.0.1:1337/v1
export LLM_CHAT_MODEL='llama3-8b-instruct'

To use Ollama locally, load a model and configure the environment variable LLM_API_BASE_URL:

ollama pull phi3
export LLM_API_BASE_URL=http://127.0.0.1:11434/v1
export LLM_CHAT_MODEL='phi3'

For LocalAI, initiate its container and adjust the environment variable LLM_API_BASE_URL:

docker run -ti -p 8080:8080 localai/localai tinyllama-chat
export LLM_API_BASE_URL=http://localhost:3928/v1

Using Managed LLM Services

To use OpenAI GPT model, configure the environment variable OPENAI_API_KEY with your API key:

export OPENAI_API_KEY="sk-yourownapikey"

To utilize other LLM services, populate the relevant environment variables as demonstrated in the following examples:

• Anyscale

export LLM_API_BASE_URL=https://api.endpoints.anyscale.com/v1
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="meta-llama/Llama-3-8b-chat-hf"

• Deep Infra

export LLM_API_BASE_URL=https://api.deepinfra.com/v1/openai
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="mistralai/Mistral-7B-Instruct-v0.1"

• Fireworks

export LLM_API_BASE_URL=https://api.fireworks.ai/inference/v1
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="accounts/fireworks/models/llama-v3-8b-instruct"

• Groq

export LLM_API_BASE_URL=https://api.groq.com/openai/v1
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="gemma-7b-it"

• Lepton

export LLM_API_BASE_URL=https://mixtral-8x7b.lepton.run/api/v1/
export LLM_API_KEY="yourownapikey"

• Novita

export LLM_API_BASE_URL=https://api.novita.ai/v3/openai
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="meta-llama/llama-3-8b-instruct"

• Octo

export LLM_API_BASE_URL=https://text.octoai.run/v1/
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="hermes-2-pro-mistral-7b"

• OpenRouter

export LLM_API_BASE_URL=https://openrouter.ai/api/v1
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="meta-llama/llama-3-8b-instruct:free"

• Together

export LLM_API_BASE_URL=https://api.together.xyz/v1
export LLM_API_KEY="yourownapikey"
export LLM_CHAT_MODEL="meta-llama/Llama-3-8b-chat-hf"