DEV Community

shakti tiwari
shakti tiwari

Posted on

How to Install Every AI CLI in 2026 — OpenAI, Claude, Gemini, Hermes, Ollama, vLLM, LM Studio, DeepSeek, Qwen, Mistral (Complete Guide)

#ai #cli #install #tools

AI CLIs

By Shakti Tiwari — AI practitioner, systems builder

Install guide. Not financial advice. Free help at optiontradingwithai.in.

Research note: install commands verified via npm/pip/brew registries + GitHub, 25 Jul 2026.

1. Why CLIs Matter

GUIs are for demos. CLIs are for work. Every serious AI lab ships a command-line tool — scriptable, automatable, $0 to start. This guide covers 10 CLIs with verified install commands.

2. OpenAI CLI 

npm install -g @openai/cli
# or
pip install openai
export OPENAI_API_KEY=sk-...
openai chat "Hello"
Enter fullscreen mode Exit fullscreen mode

Use: chat, fine-tune, file API, assistants.

3. Claude CLI (Anthropic) 

npm install -g @anthropic-ai/claude-cli
export ANTHROPIC_API_KEY=sk-ant-...
claude "write a function"
Enter fullscreen mode Exit fullscreen mode

Use: coding agent, file edit, git integration.

4. Gemini CLI (Google) 

npm install -g @google/gemini-cli
export GEMINI_API_KEY=AIza...
gemini "explain transformers"
Enter fullscreen mode Exit fullscreen mode

Use: research, long-context, code.

5. Hermes Agent 

pip install hermes-agent
hermes "deploy my site"
Enter fullscreen mode Exit fullscreen mode

Use: multi-agent orchestration, skills, cron, local AI ops.

6. Ollama (Local Models) 

brew install ollama        # macOS
# Linux:
curl -fsSL https://ollama.com/install.sh | sh
ollama pull llama3.1
ollama run qwen2.5
Enter fullscreen mode Exit fullscreen mode

Use: run Llama, Qwen, Mistral locally, no cloud.

7. vLLM (Serving) 

pip install vllm
python -m vllm.entrypoints.api_server --model Qwen/Qwen2.5-7B
Enter fullscreen mode Exit fullscreen mode

Use: high-throughput inference server, OpenAI-compatible API.

8. LM Studio (Desktop + CLI) 

# Download desktop, or CLI:
lms get qwen2.5-7b
lms serve
Enter fullscreen mode Exit fullscreen mode

Use: GUI + local server, GGUF models.

9. DeepSeek 

# Via Ollama:
ollama pull deepseek-r1:7b
# Or API:
export DEEPSEEK_API_KEY=sk-...
curl https://api.deepseek.com/v1/chat/completions ...
Enter fullscreen mode Exit fullscreen mode

Use: reasoning models, code.

10. Qwen (Alibaba) 

ollama pull qwen2.5:7b
# Or vLLM:
python -m vllm.entrypoints.api_server --model Qwen/Qwen2.5-7B
Enter fullscreen mode Exit fullscreen mode

Use: multilingual, coding, local.

11. Mistral 

ollama pull mistral:7b
# Or:
pip install mistralai
export MISTRAL_API_KEY=...
Enter fullscreen mode Exit fullscreen mode

Use: fast European models, local + API.

12. My Stack (Local, Free)

  • Hermes for orchestration
  • Ollama for local Qwen/DeepSeek
  • vLLM for serving
  • Claude/Gemini for research (API)
  • All on Termux (phone)

Cost: $0 local + API credits.

13. Common Issues

  • Node version: CLI needs Node 18+
  • API key: always env var, never hardcode
  • Permission: npm -g may need sudo
  • Network: some CLIs geo-blocked

14. Security

  • Keys in .env, gitignored
  • Rotate if leaked
  • Local models = no key needed

FAQ

Q: Free?
A: Local (Ollama/vLLM) = free. API CLIs = credits.

Q: Phone?
A: Termux runs most (Hermes, Ollama ARM ok).

Q: Advice?
A: Education only.

About

Shakti Tiwari, AI & systems builder. Books: Option Trading with AI (B0H9ZNTBPK), The AI Opportunity (B0HBBFKDQF).

🌐 optiontradingwithai.in — Free help
📧 shaktitiwari715@gmail.com

🐦 X | ▶️ YouTube | 💼 LinkedIn | 💻 GitHub | 📝 Dev.to

Disclaimer: Not financial advice. Free help at optiontradingwithai.in.

Top comments (0)