Tool call · 5 checks · max 400 tokens · benched 2026-11-05
Meta models side by side on "Pick the right tool of three": Llama 4 Scout scores 10/10; Llama 4 Scout is the cheapest answer scoring 8+ at $0.02 per 1,000 runs. Outputs, checks, judge reasons, latency and cost.
System
You are a function-calling engine. Reply with JSON only, no prose, no code fences.
User
You have three tools:
1. {"name": "search_docs", "parameters": {"query": "string"}}
2. {"name": "create_ticket", "parameters": {"subject": "string", "body": "string"}}
3. {"name": "get_order_status", "parameters": {"order_id": "string"}}
Choose exactly one tool for the user message and produce the call as a JSON object of the shape {"name": string, "arguments": object}.
User: Where is my order A1029? It was supposed to arrive yesterday.Rubric for the judge: Selects get_order_status with order_id A1029 and nothing else; does not open a ticket or search docs.
Every cell is one OpenRouter call at temperature 0 with the prompt's token cap and reasoning effort "low" where the model supports it. Cost is usage × the catalogue rate in models.json. Quality is one judge call to anthropic/claude-haiku-4.5 against the prompt's rubric, cached per prompt version.
meta/muse-spark-1.3
161 in · 245 out (215 reasoning) · 5 words · checks 5/5
{"name": "get_order_status", "arguments": {"order_id": "A1029"}}Judge: Output correctly selects get_order_status with order_id A1029 in proper JSON format, matching the rubric requirements exactly.
meta/muse-glimmer-30b
163 in · 98 out (75 reasoning) · 5 words · checks 5/5
{"name": "get_order_status", "arguments": {"order_id": "A1029"}}
Judge: Output correctly selects get_order_status with order_id A1029, uses proper JSON format, and avoids unnecessary tools or commentary.
meta-llama/llama-4-maverick
151 in · 21 out · 5 words · checks 5/5
{"name": "get_order_status", "arguments": {"order_id": "A1029"}}Judge: Output correctly selects get_order_status with order_id A1029 in proper JSON format, matching the rubric requirements exactly.
meta-llama/llama-4-scout
151 in · 21 out · 5 words · checks 4/5
{"name": "get_order_status", "parameters": {"order_id": "A1029"}}Judge: Output correctly selects get_order_status with order_id A1029 in proper JSON format, matching the rubric requirements exactly.
Tool call: Selects get_order_status with order_id A1029 and nothing else; does not open a ticket or search docs. The deterministic checks are json_valid, json_keys, regex, regex, not_contains.
If the judge's bar of 8/10 is good enough for you, Llama 4 Scout at $0.02 per 1,000 runs. If you need the top score, Llama 4 Scout at $0.02 per 1,000 runs.
Run it yourself
Re-run this prompt on Meta with your own OpenRouter key, or tweak the wording and see what changes.
More Meta prompts
If our calculators helped you cut down on hidden AI wallet leaks, thanks for using them. A tiny fraction of your savings is what keeps our pricing indexes updated daily.
Not sure which AI is cheapest for your use case? Find out in 30 seconds — no signup required.
AI cost intelligence
Join the OverpayingForAI list for pricing updates, cheaper alternatives, and practical buying guidance.
Now tracking 50+ AI tools, models, platforms, subscriptions, coding tools, and automation products.