curl

Model slugs in examples are illustrative. Confirm current availability in the live catalog and your account's /v1/models before use.

Minimal Chat Request

The smallest valid request — one user message, no system prompt:

curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kimi-k2.7-code",
    "messages": [
      {"role": "user", "content": "What is a context window?"}
    ]
  }'

With a System Prompt and Parameters

Add a system message and tune temperature and max_tokens:

curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-pro",
    "messages": [
      {"role": "system", "content": "You are a senior software engineer. Be direct and precise."},
      {"role": "user",   "content": "What does Big-O notation tell you about an algorithm?"}
    ],
    "temperature": 0.3,
    "max_tokens": 200
  }'

Streaming Output

Set stream: true to receive the response incrementally as server-sent events:

curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3.1-flash-lite",
    "messages": [{"role": "user", "content": "Write a haiku about distributed systems."}],
    "stream": true
  }'

Each line in the response looks like data: {...}. The final line is data: [DONE].

Calling Different Models

# MiniMax M3 — best for long-context work (1M tokens)
curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "minimax-m3", "messages": [{"role": "user", "content": "Explain transformer attention."}]}'

# DeepSeek — cost-efficient reasoning
curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek-v4-pro", "messages": [{"role": "user", "content": "Spot the bug in this SQL query."}]}'

# Z.AI GLM 5.3 Flash — multimodal, 1M context, always-on thinking
curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "glm-5.3-flash", "messages": [{"role": "user", "content": "Review this implementation plan."}]}'

List All Available Models

curl https://api.llmai.dev/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

Practical Tips

Keep your key out of shell history — store it in an environment variable:

export LLMAI_KEY="sk-llmaai-your-key-here"

curl https://api.llmai.dev/v1/chat/completions \
  -H "Authorization: Bearer $LLMAI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "kimi-k2.7-code", "messages": [{"role": "user", "content": "Hello!"}]}'

Extract just the reply text with jq:

curl ... | jq -r '.choices[0].message.content'

Check token usage from the response:

curl ... | jq '.usage'