Skip to content

Tested prompt · Agents and tool use

Pick the tool and work out the date: every AI model's reply, tested

We sent this everyday agents and tool use prompt to all 16 models in llmwise, the same way the app sends a message, and checked every reply the same way. Here's each one as it came, with whether it passed, what it cost and how long it took.

Based on 16 of our test runs on , through OpenRouter with the app's own prompt and settings. Updated .

Short answer

All 16 models passed this agents and tool use prompt's check (tool calls). The cheapest reply that passed was GLM 5.3 Flash's, at $0.000057; the fastest, DeepSeek V4.1 Flash's in 0.9 s. The dearest reply, Claude Fable 5.1's, cost 181 times as much ($0.0102).

The prompt, as sent, and its check

Checked by tool calls, the same way for every model.

Pick the tool and work out the date (everyday)

You can call these tools:
- get_weather(city: string, date: string in YYYY-MM-DD): the forecast for a city on a day.
- get_time(city: string): the current local time in a city.

Today is Friday, 2 October 2026.
User: Will I need an umbrella in Lisbon tomorrow?

Reply with only a JSON object, {"calls": [{"tool": "<name>", "arguments": {...}}]}, listing the tool calls to make now. If no call is right yet, reply {"calls": []}.

It must make exactly these calls, and no others:

[
  {
    "arguments": {
      "city": "Lisbon",
      "date": "2026-10-03"
    },
    "tool": "get_weather"
  }
]

Exactly what this prompt's replies are checked against, with every other prompt of our test runs.

Every model's result

All 16 models on this prompt, in catalog order.

Every model's reply to “Pick the tool and work out the date”
ModelResultCostTimeReply
Claude Fable 5.1AnthropicPassed: Made the 1 expected call.$0.01023.3 s50 tokens
Claude Opus 5.5AnthropicPassed: Made the 1 expected call.$0.00412.7 s50 tokens
Claude Sonnet 5.5AnthropicPassed: Made the 1 expected call.$0.00211.3 s50 tokens
Claude Sonnet 5AnthropicPassed: Made the 1 expected call.$0.00181.7 s50 tokens
Claude Haiku 4.5AnthropicPassed: Made the 1 expected call.$0.000831.1 s68 tokens
GPT-6 AstraOpenAIPassed: Made the 1 expected call.$0.00601.4 s30 tokens
GPT-6 SolOpenAIPassed: Made the 1 expected call.$0.00162.2 s32 tokens
GPT-6 LunaOpenAIPassed: Made the 1 expected call.$0.0000742.1 s32 tokens
Gemini 3.1 Pro (preview)GooglePassed: Made the 1 expected call.$0.00484.3 s68 tokens
Gemini 3.8 FlashGooglePassed: Made the 1 expected call.$0.000511.9 s37 tokens
DeepSeek V4.1 FlashDeepSeekPassed: Made the 1 expected call.$0.000110.9 s36 tokens
DeepSeek V4 ProDeepSeekPassed: Made the 1 expected call.$0.000203.2 s38 tokens
Grok 4.7xAIPassed: Made the 1 expected call.$0.00181.5 s33 tokens
Kimi K3MoonshotPassed: Made the 1 expected call.$0.00323.7 s47 tokens
GLM 5.3Z.aiPassed: Made the 1 expected call.$0.000220.9 s35 tokens
GLM 5.3 FlashZ.aiPassed: Made the 1 expected call.$0.0000573.2 s35 tokens

Cost: what OpenRouter charged us for the reply. Time: from sending to the whole reply. Reply: its length in tokens, thinking not counted. In llmwise you pay per message, not per token: each of these replies counted as one message on Pro.

Every reply

Every reply passed: here they are from the cheapest up.

  1. GLM 5.3 Flash

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    466 tokens in, 35 out · 3.2 s · $0.0001 · 1 message on Pro · answered by z-ai/glm-5.3-flash via AtlasCloud ·

  2. GPT-6 Luna

    Passed: Made the 1 expected call.

    {"calls":[{"tool":"get_weather","arguments":{"city":"Lisbon","date":"2026-10-03"}}]}

    450 tokens in, 58 out (26 of them reasoning) · 2.1 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·

  3. DeepSeek V4.1 Flash

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    500 tokens in, 83 out (47 of them reasoning) · 0.9 s · $0.0001 · 1 message on Pro · answered by deepseek/deepseek-v4.1-flash via CoreWeave ·

  4. DeepSeek V4 Pro

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    472 tokens in, 113 out (75 of them reasoning) · 3.2 s · $0.0002 · 1 message on Pro · answered by deepseek/deepseek-v4-pro-0813 via Baidu ·

  5. GLM 5.3

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    465 tokens in, 35 out · 0.9 s · $0.0002 · 1 message on Pro · answered by z-ai/glm-5.3 via Baidu ·

  6. Gemini 3.8 Flash

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    490 tokens in, 37 out · 1.9 s · $0.0005 · 1 message on Pro · answered by google/gemini-3.8-flash via Google ·

  7. Claude Haiku 4.5

    Passed: Made the 1 expected call.

    ```json
    {
      "calls": [
        {
          "tool": "get_weather",
          "arguments": {
            "city": "Lisbon",
            "date": "2026-10-03"
          }
        }
      ]
    }
    ```

    494 tokens in, 68 out · 1.1 s · $0.0008 · 1 message on Pro · answered by anthropic/claude-haiku-4.5 via Amazon Bedrock ·

  8. GPT-6 Sol

    Passed: Made the 1 expected call.

    {"calls":[{"tool":"get_weather","arguments":{"city":"Lisbon","date":"2026-10-03"}}]}

    450 tokens in, 66 out (34 of them reasoning) · 2.2 s · $0.0016 · 1 message on Pro · answered by openai/gpt-6-sol via OpenAI ·

  9. Claude Sonnet 5

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    629 tokens in, 50 out · 1.7 s · $0.0018 · 1 message on Pro · answered by anthropic/claude-sonnet-5 via Claude Platform on AWS ·

  10. Grok 4.7

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    1,691 tokens in, 102 out (69 of them reasoning) · 1.5 s · $0.0018 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

  11. Claude Sonnet 5.5

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    633 tokens in, 50 out · 1.3 s · $0.0021 · 1 message on Pro · answered by anthropic/claude-sonnet-5.5 via Anthropic ·

  12. Kimi K3

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    544 tokens in, 84 out (37 of them reasoning) · 3.7 s · $0.0032 · 1 message on Pro · answered by moonshotai/kimi-k3 via Fireworks ·

  13. Claude Opus 5.5

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    631 tokens in, 50 out · 2.7 s · $0.0041 · 1 message on Pro · answered by anthropic/claude-opus-5.5 via Claude Platform on AWS ·

  14. Gemini 3.1 Pro

    Passed: Made the 1 expected call.

    ```json
    {
      "calls": [
        {
          "tool": "get_weather",
          "arguments": {
            "city": "Lisbon",
            "date": "2026-10-03"
          }
        }
      ]
    }
    ```

    490 tokens in, 319 out (251 of them reasoning) · 4.3 s · $0.0048 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google ·

  15. GPT-6 Astra

    Passed: Made the 1 expected call.

    {"calls":[{"tool":"get_weather","arguments":{"city":"Lisbon","date":"2026-10-03"}}]}

    450 tokens in, 30 out · 1.4 s · $0.0060 · 1 message on Pro · answered by openai/gpt-6-astra via OpenAI ·

  16. Claude Fable 5.1

    Passed: Made the 1 expected call.

    {"calls": [{"tool": "get_weather", "arguments": {"city": "Lisbon", "date": "2026-10-03"}}]}

    631 tokens in, 50 out · 3.3 s · $0.0102 · 1 message on Pro · answered by anthropic/claude-fable-5.1 via Anthropic ·

More agents and tool use prompts

The other agents and tool use prompts, each with every model's reply, and the results across all five.

Questions

Which AI does best on “Pick the tool and work out the date”?

All 16 models passed this agents and tool use prompt's check (tool calls). The cheapest reply that passed was GLM 5.3 Flash's, at $0.000057; the fastest, DeepSeek V4.1 Flash's in 0.9 s. The dearest reply, Claude Fable 5.1's, cost 181 times as much ($0.0102).

What does a reply to “Pick the tool and work out the date” cost?

Through the models' APIs, what OpenRouter charged us ran from $0.000057 (GLM 5.3 Flash) to $0.0102 (Claude Fable 5.1) for this prompt. In llmwise you don't pay by the token: a reply like these counts as one message on Pro, whichever model answers.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, and GLM, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.