Chat completions

​

Example

require 'ai_stats_sdk'

client = AIStatsSdk::AIStats.new(api_key: ENV.fetch("AI_STATS_API_KEY"))

response = client.create_chat_completion(
  "model" => "openai/gpt-4o-mini",
  messages: [
    { "role" => "user", "content" => "Write a limerick about lighthouses." }
  ],
  "temperature" => 0.5
)

​

Key parameters

• model (required): Target model id.
• messages (required): Ordered messages with roles user/system/assistant/tool; content as strings.
• Sampling: temperature (0–2), top_p (0–1), top_k (>=1), seed (int, optional).
• Length/penalties: max_output_tokens (int), presence_penalty and frequency_penalty (-2 to 2), stop (string|string[]).
• Tools: tools (definitions), tool_choice (auto/none/specific tool), max_tool_calls (int), parallel_tool_calls (bool).
• Logprobs: logprobs (bool), top_logprobs (0–20).
• Output: response_format (json/text), metadata (object passthrough), stream (bool), service_tier.
• Gateway extras: usage (bool to request usage), meta (bool to include meta block).

​

Returns