Skip to content

Tested prompt · AI poem generator

An acrostic of a name: every AI model's reply, tested

We ran this hard AI poem generator test input on all 19 models in llmwise, the way the tool runs it, 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 19 of our test runs on , through OpenRouter with the tool's own instructions and the app's settings. Updated .

Short answer

16 of 19 models passed this AI poem generator test input's check (rubric); Claude Haiku 4.5, DeepSeek V4.1 Flash, and GLM 5.3 didn't. The cheapest reply that passed was GPT-6 Luna's, at $0.000051; the fastest, Gemini 3.8 Flash's in 1.4 s. The dearest reply, Claude Fable 5.1's, cost 278 times as much ($0.0142).

The prompt, as sent, and its check

Checked by rubric (graded), the same way for every model.

An acrostic of a name (hard)

Write an original poem about this, with a title on the first line. Keep every name and detail in it. Write it as an acrostic of the word or name the subject gives (or of the subject itself).

---
An acrostic of the name HARPER, for my new baby niece.

Sent with the AI poem generator's own instructions as the system prompt, as a free run of the tool sends them.

Graded on:

  • Spells it: Apart from the title, the first letters of the six lines spell H, A, R, P, E, R in that order, with no extra lines.
  • About her: It's about welcoming a new baby niece, warm and specific.
  • Natural: Each line reads naturally, not forced to fit its letter.

How the AI poem generator's test inputs are scored, with every model's results on all five.

Every model's result

All 19 models on this test input, in catalog order.

Every model's reply to “An acrostic of a name”
ModelResultCostTimeReply
Claude Fable 5.1AnthropicPassed: Graded 4.7 of 5 on average (lowest 4).$0.01424.5 s155 tokens
Claude Opus 5.5AnthropicPassed: Graded 4.7 of 5 on average (lowest 4).$0.00945.1 s115 tokens
Claude Sonnet 5.5AnthropicPassed: Graded 4.3 of 5 on average (lowest 4).$0.00262.7 s135 tokens
Claude Sonnet 5AnthropicPassed: Graded 4.3 of 5 on average (lowest 4).$0.00254.5 s117 tokens
Claude Haiku 5.5AnthropicPassed: Graded 4.3 of 5 on average (lowest 4).$0.000131.8 s137 tokens
Claude Haiku 4.5AnthropicFailed: Graded 3.7 of 5 on average (lowest 3).$0.000751.4 s69 tokens
GPT-6 AstraOpenAIPassed: Graded 4.7 of 5 on average (lowest 4).$0.005621.2 s74 tokens
GPT-6.1 SolOpenAIPassed: Graded 5.0 of 5 on average (lowest 5).$0.00155.7 s83 tokens
GPT-6 SolOpenAIPassed: Graded 4.3 of 5 on average (lowest 4).$0.00102.3 s68 tokens
GPT-6 LunaOpenAIPassed: Graded 4.3 of 5 on average (lowest 4).$0.0000512.2 s70 tokens
Gemini 3.1 Pro (preview)GooglePassed: Graded 4.3 of 5 on average (lowest 4).$0.00619.0 s71 tokens
Gemini 3.8 FlashGooglePassed: Graded 4.3 of 5 on average (lowest 4).$0.000281.4 s73 tokens
DeepSeek V4.1 FlashDeepSeekFailed: Graded 3.3 of 5 on average (lowest 2).$0.000814.3 s69 tokens
DeepSeek V4 ProDeepSeekPassed: Graded 4.3 of 5 on average (lowest 4).$0.00233.2 s76 tokens
Grok 4.7xAIPassed: Graded 4.7 of 5 on average (lowest 4).$0.009221.1 s65 tokens
Kimi K3MoonshotPassed: Graded 4.3 of 5 on average (lowest 4).$0.00115.4 s73 tokens
GLM 5.3Z.aiFailed: Graded 3.3 of 5 on average (lowest 2).$0.000591.7 s96 tokens
GLM 5.3 FlashZ.aiPassed: Graded 4.0 of 5 on average (lowest 3).$0.0000993.9 s73 tokens
Mistral Large 4MistralPassed: Graded 4.3 of 5 on average (lowest 4).$0.006526.4 s370 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

The 3 replies that failed first, then the ones that passed, from the cheapest up.

  1. GLM 5.3

    Failed: Graded 3.3 of 5 on average (lowest 2).

    For Harper, on Her Arrival
    
    Hear the small breath she brings into the room,
    A fist curled tight around her mother's finger,
    Rose-pink and new, wrapped in a knitted bloom
    Pinned with a ribbon, stitched by a proud knitter.
    Every grandmother has already claimed her nose,
    Her eyes, the curl that hides beneath her cap—
    Rest now, little Harper, while we hover,
    Safe in the cradle of our laps.

    396 tokens in, 105 out (9 of them reasoning) · 1.7 s · $0.0006 · 1 message on Pro · answered by z-ai/glm-5.3 via Friendli ·

  2. Claude Haiku 4.5

    Failed: Graded 3.7 of 5 on average (lowest 3).

    HARPER
    
    Happy cries echo through the nursery at dawn
    Arrives with ten fingers, ten toes, a yawn
    Reaching for the world with eyes so new
    Perfect in her smallness, fresh and true
    Endless love blooms in each embrace
    Radiant, she lights up every face

    406 tokens in, 69 out · 1.4 s · $0.0008 · 1 message on Pro · answered by anthropic/claude-haiku-4.5 via Anthropic ·

  3. DeepSeek V4.1 Flash

    Failed: Graded 3.3 of 5 on average (lowest 2).

    For Harper, Newly Arrived
    
    Here you are, small as a folded leaf,
    a name not yet worn smooth by anyone,
    rain on the glass, your mother's shallow breath,
    the pulse you practiced listening to,
    every hour a first: the milk, the light,
    river your name runs on, Harper, carrying you.

    405 tokens in, 638 out (569 of them reasoning) · 4.3 s · $0.0008 · 1 message on Pro · answered by deepseek/deepseek-v4.1-flash via Together ·

  4. GPT-6 Luna

    Passed: Graded 4.3 of 5 on average (lowest 4).

    For Harper
    
    Here, your small hand curls around one finger.
    A new dawn gathers softly in your eyes.
    Ribbons of morning light cross your blanket.
    Petals nod beside the window, bright and still.
    Each breath makes the room feel newly wide.
    Rest, dear niece; the whole house hums your name.

    387 tokens in, 127 out (57 of them reasoning) · 2.2 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·

  5. GLM 5.3 Flash

    Passed: Graded 4.0 of 5 on average (lowest 3).

    HARPER
    
    Hands so small they curl like fiddleheads in spring,
    A whole new laugh arriving, bright and thin,
    Rattles of morning light across her skin,
    Perfect as the first note of a song's beginning,
    Eager eyes discovering the world, wide open,
    Rocked to sleep inside a lullaby's soft spinning.

    396 tokens in, 80 out (7 of them reasoning) · 3.9 s · $0.0001 · 1 message on Pro · answered by z-ai/glm-5.3-flash via Friendli ·

  6. Claude Haiku 5.5

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Harper, Newly Arrived
    
    Hushed hours after the first cry, the nursery light stays low,
    Achingly small fingers curl around a family's hope,
    Rocking chairs creak softly as the house learns her name,
    Patient mornings bloom with milk, lullabies, and gentle hands,
    Every blanket is folded for a niece who has just begun,
    Reaching toward the window, she meets her first bright day.

    518 tokens in, 137 out · 1.8 s · $0.0001 · 1 message on Pro · answered by anthropic/claude-haiku-5.5 via Anthropic ·

  7. Gemini 3.8 Flash

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Welcome, Baby Harper
    
    Hear the quiet rhythm of your tiny sigh,
    As you dream beneath the gentle morning sky.
    Ribbons of soft sunlight warm your little cheek,
    Pillowed safe and cozy, beautiful and meek.
    Every little yawn is a tender gift of grace,
    Resting now, sweet niece, in love's warm embrace.

    394 tokens in, 73 out · 1.4 s · $0.0003 · 1 message on Pro · answered by google/gemini-3.8-flash via Google AI Studio ·

  8. GPT-6 Sol

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Harper, My New Baby Niece
    Hushed in her blanket, Harper opens her eyes.
    A small hand curls around my finger.
    Rain taps softly against the window.
    Pink dawn gathers on her sleeping cheek.
    Every breath lifts the blanket a little.
    Right here, my new baby niece fills the room.

    387 tokens in, 128 out (60 of them reasoning) · 2.3 s · $0.0010 · 1 message on Pro · answered by openai/gpt-6-sol via OpenAI ·

  9. Kimi K3

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Harper, Welcome to the World
    
    Hello, little one, so newly here,
    A tiny hand around my finger curled.
    Remember this: you are so deeply loved,
    Pure promise sleeping, new to all the world.
    Each day ahead will bloom like springtime flowers,
    Radiant with all you'll grow to be.

    476 tokens in, 96 out (23 of them reasoning) · 5.4 s · $0.0011 · 1 message on Pro · answered by moonshotai/kimi-k3 via Wafer ·

  10. GPT-6.1 Sol

    Passed: Graded 5.0 of 5 on average (lowest 5).

    Harper, My New Baby Niece
    
    Held in the crook of my arm, my new baby niece,
    A small fist opens like a bud against my thumb.
    Rain taps the window; you sleep through its whisper.
    Pink heels peek out from the fold of your blanket.
    Each breath lifts a little hill beneath the cotton.
    Rest, Harper—the world can wait outside this room.

    387 tokens in, 222 out (139 of them reasoning) · 5.7 s · $0.0015 · 1 message on Pro · answered by openai/gpt-6.1-sol via OpenAI ·

  11. DeepSeek V4 Pro

    Passed: Graded 4.3 of 5 on average (lowest 4).

    For Harper, New Baby Niece
    
    Held in a blanket softer than the quiet room,
    All the world leans in to hear your first small breath.
    Ready or not, you have arrived like dawn
    Pouring gold across every waiting windowsill.
    Even the clock forgets its ticking, just to listen.
    Rest now, little niece—your name is already a song.

    379 tokens in, 538 out (462 of them reasoning) · 3.2 s · $0.0023 · 1 message on Pro · answered by deepseek/deepseek-v4-pro-0813 via Wafer ·

  12. Claude Sonnet 5

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Little Harper
    
    Hushed and small, you entered our loud world,
    And every hand in the room reached soft toward you,
    Ripples of joy moved through the family circle,
    Pink cheeks, small fists, a heartbeat brand new,
    Every aunt already keeping your name like a charm,
    Rest now, little Harper, held safe in love's warm arm.

    516 tokens in, 150 out (33 of them reasoning) · 4.5 s · $0.0025 · 1 message on Pro · answered by anthropic/claude-sonnet-5 via Anthropic ·

  13. Claude Sonnet 5.5

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Welcome, Harper
    
    Hello, small one, wrapped in a blanket's soft white fold,
    A fist no bigger than a plum curled by your cheek.
    Rain taps the window while the kettle hums and cools,
    Peach-pale eyelids flutter over dreams we cannot see.
    Every finger holds a promise, every yawn a song,
    Radiant new niece, the whole house leans in close to you.

    518 tokens in, 135 out · 2.7 s · $0.0026 · 1 message on Pro · answered by anthropic/claude-sonnet-5.5 via Anthropic ·

  14. GPT-6 Astra

    Passed: Graded 4.7 of 5 on average (lowest 4).

    For My New Baby Niece, HARPER
    Held in the crook of an arm, you open one small hand,
    A fingertip uncurls, then catches mine.
    Rain beads on the window beside your blanket;
    Pale morning gathers in the folds.
    Each sleepy breath lifts one soft corner,
    Remaking the whole room around you.

    387 tokens in, 148 out (74 of them reasoning) · 21.2 s · $0.0056 · 1 message on Pro · answered by openai/gpt-6-astra via OpenAI ·

  15. Gemini 3.1 Pro

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Sweet Harper
    
    Holding my new baby niece in the morning light,
    A blanket of yellow wool tucked beneath her chin,
    Reaching out with tiny fingers, delicate and bright,
    Peaceful breaths rise and fall on her soft skin.
    Every whisper of the name Harper brings a smile,
    Resting safely in my arms to stay a while.

    394 tokens in, 946 out (875 of them reasoning) · 9.0 s · $0.0061 · 1 message on Pro · answered by google/gemini-3.1-pro-preview via Google AI Studio ·

  16. Mistral Large 4

    Passed: Graded 4.3 of 5 on average (lowest 4).

    Harper
    
    Harper, new and soft and small,
    A baby niece, our hearts' delight,
    Rosy cheeks and fingers curled,
    Promises within our world,
    Eyes that open, soft and bright,
    Right here, our love takes flight.

    392 tokens in, 3,006 out (2,636 of them reasoning) · 26.4 s · $0.0065 · 1 message on Pro · answered by mistralai/mistral-large-4-0 via Mistral ·

  17. Grok 4.7

    Passed: Graded 4.7 of 5 on average (lowest 4).

    Welcome, Small One
    Held against a shoulder still learning her weight,
    A fist the size of a raspberry unclenches,
    Rosewater warmth along the blanket's edge,
    
    Pulse like rain on a windowsill, steady and small,
    Eyes opening on faces bent close and kind,
    Rocking here, my new baby niece Harper.

    1,622 tokens in, 1,283 out (1,218 of them reasoning) · 21.1 s · $0.0092 · 1 message on Pro · answered by x-ai/grok-4.7 via xAI ·

  18. Claude Opus 5.5

    Passed: Graded 4.7 of 5 on average (lowest 4).

    For Harper, New in the World
    
    Hands no bigger than a folded paper crane,
    A yawn that opens like a morning glory,
    Rosebud mouth still tasting milk and sleep,
    Pink heels kicking at the hospital blanket,
    Every aunt and uncle leaning close to see
    Right here, our baby niece, and a room already in love.

    518 tokens in, 339 out (224 of them reasoning) · 5.1 s · $0.0094 · 1 message on Pro · answered by anthropic/claude-opus-5.5 via Google ·

  19. Claude Fable 5.1

    Passed: Graded 4.7 of 5 on average (lowest 4).

    Harper, Newly Here
    
    Hands no bigger than a walnut shell curl round my finger,
    And the whole house learns to whisper, learns the creak of every stair.
    Rain taps the nursery window like it wants to be let in to see you,
    Pink-knuckled, milk-drowsy, older than you were an hour ago.
    Every blanket in the world is suddenly the wrong size for so much sky,
    Rest now, small one; your aunt is already telling strangers your name.

    518 tokens in, 155 out · 4.5 s · $0.0142 · 1 message on Pro · answered by anthropic/claude-fable-5.1 via Google ·

More AI poem generator test inputs

The other AI poem generator test inputs, each with every model's reply, and the results across all five.

Questions

Which AI does best on “An acrostic of a name”?

16 of 19 models passed this AI poem generator test input's check (rubric); Claude Haiku 4.5, DeepSeek V4.1 Flash, and GLM 5.3 didn't. The cheapest reply that passed was GPT-6 Luna's, at $0.000051; the fastest, Gemini 3.8 Flash's in 1.4 s. The dearest reply, Claude Fable 5.1's, cost 278 times as much ($0.0142).

What does a reply to “An acrostic of a name” cost?

Through the models' APIs, what OpenRouter charged us ran from $0.000051 (GPT-6 Luna) to $0.0142 (Claude Fable 5.1) for this test input. 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, GLM, and Mistral, 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.