Use free AI from your shell.

Amazon Q Developer Free includes 50 agentic requests per month and 1,000 lines of Java transformation per month.

q whoami
q chat

Amazon Q's Free tier includes 50 agentic requests per month. AWS documentation currently notes that the Q CLI is transitioning to Kiro CLI.

Cloudflare Workers AI includes 10,000 Neurons of AI inference per day at no charge on Workers Free.

export CLOUDFLARE_ACCOUNT_ID='...'
export CLOUDFLARE_API_TOKEN='...'
curl -sS --request POST \
  --url "https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/chat/completions" \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{"model":"@cf/meta/llama-3.1-8b-instruct","messages":[{"role":"user","content":"Hello from the terminal."}]}'

Cloudflare measures this allocation in Neurons, not a fixed token count. Use a model allowed on Workers Free.

Cohere evaluation/trial keys are free and limited to 1,000 API calls per month, with endpoint-specific per-minute limits.

export CO_API_KEY='...'
curl -sS --request POST \
  --url https://api.cohere.ai/v2/chat \
  --header "Authorization: bearer $CO_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{"model":"command-a-plus-05-2026","messages":[{"role":"user","content":"Hello from the terminal."}],"max_tokens":128}'

Cohere's free evaluation key is limited by calls, not by a published token allowance.

Eligible Gemini models have free-of-charge input and output tokens within model-specific rate limits.

export GEMINI_API_KEY='...'
curl -sS -X POST "https://generativelanguage.googleapis.com/v1beta/interactions" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-3.8-flash","input":"Hello from the terminal."}'

Use a model that is currently marked Free Tier in Google AI Studio. Quotas are model and project specific.

Groq publishes a Free-plan table. Listed GPT-OSS and Qwen text models currently show 200,000 tokens and 1,000 requests per day.

export GROQ_API_KEY='...'
curl -sS https://api.groq.com/openai/v1/chat/completions \
  -H "Authorization: Bearer $GROQ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-oss-20b","messages":[{"role":"user","content":"Hello from the terminal."}],"max_tokens":128}'

Groq exposes an OpenAI-compatible endpoint. The exact Free-plan limits are shown per model in the linked rate-limit table.

Every free Hugging Face user receives a small monthly credit for routed Inference Providers requests.

export HF_TOKEN='...'
curl -sS https://router.huggingface.co/v1/chat/completions \
  -H "Authorization: Bearer $HF_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-oss-120b:fastest","messages":[{"role":"user","content":"Hello from the terminal."}]}'

The free account credit is $0.10 per month and is consumed by routed provider usage; the amount is subject to change.

Kiro's perpetual Free tier includes 50 credits for prompts, specs, tasks, and hooks.

kiro-cli chat

Authenticate with Kiro CLI as documented, then start an interactive chat. Kiro CLI consumes the Free tier's monthly credits.

OpenCode Zen lists seven models with free input and output pricing. No fixed token quota is published.

npm install -g opencode-ai
cd /path/to/project
opencode
# In OpenCode: /connect

OpenCode is the terminal client. Connect a provider key in /connect; Zen itself does not publish a fixed free-token quota.

Use free model variants through OpenRouter chat or its API, with a published Free-plan limit of 50 requests per day.

export OPENROUTER_API_KEY='...'
curl -sS https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openrouter/free","messages":[{"role":"user","content":"Hello from the terminal."}],"max_tokens":128}'

OpenRouter exposes the free-model router through an OpenAI-compatible chat endpoint.