eval

type Comparison struct {
	Models map[string]Summary `json:"models"`
}

type Evaluator struct{}

type RunMetrics struct {
	Model            string        `json:"model"`
	Duration         time.Duration `json:"duration"`
	PromptTokens     int           `json:"prompt_tokens"`
	CompletionTokens int           `json:"completion_tokens"`
	TotalTokens      int           `json:"total_tokens"`
	ToolCalls        int           `json:"tool_calls"`
	Success          bool          `json:"success"`
	Error            string        `json:"error,omitempty"`
	Timestamp        time.Time     `json:"timestamp"`
}

type Scenario struct {
	Input string
	// ExpectedContains checks if model output includes this substring (minimal oracle)
	ExpectedContains string
}

type Summary struct {
	Runs            int                `json:"runs"`
	Successes       int                `json:"successes"`
	Failures        int                `json:"failures"`
	SuccessRate     float64            `json:"success_rate"`
	AvgLatencyMS    float64            `json:"avg_latency_ms"`
	AvgPromptTokens float64            `json:"avg_prompt_tokens"`
	AvgCompletion   float64            `json:"avg_completion_tokens"`
	AvgTotalTokens  float64            `json:"avg_total_tokens"`
	TotalToolCalls  int                `json:"total_tool_calls"`
	ByModel         map[string]Summary `json:"by_model,omitempty"`
}