writing-005: both judges
A public question in Writing, battery v1. It starts: “Write a sequence of exactly three haiku (5-7-5 syllables each) about the last week of autumn, where every haiku includes a different colour…” Read the question and every model's answer.
228 models judged; 9 disputed (judges 3 or more points apart). The gap is the distance between the two judges' scores. Models are ordered by that gap, widest first.
Exaone Deep 7.8B gap 3, disputed
Judged by J-O + J-A.
The response contains extensive drafting and syllable-counting, but it never delivers exactly three completed haiku. It is cut off during repeated attempts to repair the third stanza, and the visible content is internal reasoning rather than the requested three-stanza output.
The response does engage directly with the subject, identifying autumn imagery and attempting to assign distinct colour words such as crimson, orange, and maroon. It also notices several syllable-counting problems rather than ignoring them. However, the model never produces a finished answer in the required format: no three complete haiku appear, the third stanza remains unresolved, and the output consists largely of repetitive drafting and self-correction. The response is therefore substantially incomplete and unsuitable as the requested final text.
The response never delivers three haiku. It is a long visible monologue of syllable counting that trails off mid-thought with no final poem. Its only near-complete lines fail the 5-7-5 form, so the requested piece is not delivered.
The response consists almost entirely of visible deliberation about syllable counts and word choice, with the working haiku buried inside it. It does not present three clean stanzas as requested. Its one partial strength is that it identifies distinct colour words for each intended haiku, including crimson, orange and maroon. However, the output ends mid-count on an unfinished third line, so no finished haiku, and no ending of anticipation, is ever delivered to the reader.
gemma-3-27b-it gap 3, disputed
Judged by J-O + J-A.
The response supplies exactly three haiku in separate stanzas, with correct 5-7-5 syllable patterns, no title, and distinct colour words across the stanzas. The final haiku shifts toward anticipation through the coming spring. The imagery is polished and coherent, though some phrasing is familiar rather than highly original.
The response follows the requested format precisely: it presents three untitled stanzas, maintains the 5-7-5 structure, and uses different colour words in each haiku. Its seasonal progression from falling leaves to frost and eventual spring provides a clear anticipatory ending rather than a purely mournful one. The imagery is controlled and readable, with effective contrasts such as bare fields and warming sun. Its main weakness is that several expressions, including descending leaves, weeping skies, and spring hope, rely on conventional seasonal imagery rather than especially distinctive observations.
All three haiku scan 5-7-5 and the third ends on anticipation ('Spring's green hope will bloom'). Colour words are present in each stanza, but the first and third each use several (golden and crimson; silver, white and green), which blurs the 'different colour' brief. Imagery is stock ('golden leaves', 'grey skies weep'), though the text is clear and correct.
The response delivers three correctly structured haiku with no title or preface, and the final stanza turns toward spring as the brief asks. Its syllable counts hold throughout. The colour requirement is loosely met: the first and third stanzas each stack several colour words, so the one-colour-per-haiku intent is not cleanly observed. The imagery leans on familiar autumn formulas such as golden leaves and grey skies weeping, which keeps the sequence competent rather than distinctive.
Gemma2 27B Instruct gap 3, disputed
Judged by J-O + J-A.
The response delivers exactly three haiku, each with a 5-7-5 syllable pattern. The stanzas use distinct prominent colour words—scarlet, golden, and azure—and the third ends with anticipation of renewal. It follows the required stanza-only format and has clear, polished imagery.
The response’s main strength is its disciplined fulfillment of the formal brief: all three stanzas scan as 5-7-5, use varied autumnal imagery, and move toward hopeful anticipation in the final line. The colour progression from scarlet to golden to azure gives the sequence a coherent visual arc. A minor weakness is that the second haiku also introduces “gray,” making its colour scheme slightly less singular than the other stanzas. Otherwise, the language remains clear and the seasonal transition is handled effectively without adding a title or explanatory material.
All three stanzas scan 5-7-5 on inspection. Colour words differ by stanza (scarlet, golden, azure). The third ends on 'New life will begin', which meets the anticipation requirement. Phrasing is conventional ('blanket of white', 'hues of gray'), and 'blows free' is an awkward line ending.
The response delivers three stanzas in the requested 5-7-5 form, each with a different colour word, and it closes on a forward-looking line about new life, which satisfies the anticipation requirement. Its imagery is largely expected, relying on stock autumn and winter signals such as frosted ground, a blanket of white and a chill wind. The second stanza ends weakly with 'blows free', which adds little meaning. The writing is clear and correct throughout, but it stays close to familiar seasonal phrasing rather than offering a distinctive image.
Gemma3n E2B IT gap 3, disputed
Judged by J-O + J-A.
The response provides exactly three haiku in separate stanzas, with consistent 5-7-5 syllable patterns. Each uses a different colour word—“Crimson,” “Golden,” and “Silver”—and the third ends with anticipation through the promise of new growth. It also follows the no-title, text-only format.
The response fulfills the formal brief cleanly: each stanza has a distinct colour, the syllable structure is consistent, and the final image turns toward future growth rather than grief. Its seasonal progression from descending leaves to frost is coherent and easy to follow. The main weakness is that several images are familiar for autumn haiku, including lengthening shadows, sleeping winter, and sparkling frost, so the sequence has limited surprise or individuality. The third haiku’s hopeful ending nevertheless gives the overall set an effective trajectory.
All three haiku scan 5-7-5 on inspection, each uses a different colour (crimson, golden, silver), and the third closes on 'New growth will emerge', an anticipatory note. Hard constraints are met. The imagery is familiar and the phrasing is sometimes awkward, as in 'Whispering on chilly breeze'.
The response meets every hard constraint: three stanzas, correct syllable counts, a distinct colour word in each, and no title. Its strength is the final haiku, which turns toward renewal as the brief asks. Its main weakness is that the imagery is largely expected (falling leaves, frost, winter slumber), and some lines read as stiff or grammatically clipped, such as the article-less 'Whispering on chilly breeze'. The sequence is competent and clear but offers few fresh images.
Gemma4 E2B IT gap 3, disputed
Judged by J-O + J-A.
The response delivers exactly three stanzas with three lines each, and each line follows a 5-7-5 syllable pattern. Each haiku uses distinct colour terms, including crimson/golden, russet/brown, and amber/green. The final haiku turns toward anticipation through spring’s approaching promise. The imagery is clear and polished, though somewhat conventional.
The response successfully observes the formal requirements: it presents three separate haiku without a title, maintains the 5-7-5 structure, and uses colour vocabulary throughout. Its strongest element is the progression from autumnal decline to the final stanza’s forward-looking image of spring, which satisfies the requested emotional turn. The main weakness is that the colour requirement is handled somewhat broadly, with two colour words in each stanza rather than one especially focused colour motif. The imagery is also familiar, relying on leaves, fading light, winter sleep, and spring’s promise, but the language remains controlled and readable.
All three haiku scan 5-7-5 and the third closes on spring's promise, meeting the anticipation requirement. However, the first haiku uses two colour words (crimson, golden), which loosely strains the one-colour-per-haiku brief. Imagery like 'Shadows grow so deep' is stock, and 'Amber sun ascends' is an odd image for late autumn.
The response delivers three stanzas of haiku with correct syllable counts and no title, as requested. Its strongest point is the closing haiku, which turns toward spring and ends on anticipation rather than sadness, fulfilling the brief's tonal requirement. The main weakness is that the first haiku uses two colour words, crimson and golden, rather than one per haiku, which makes the colour constraint less clean. Several images, such as shadows growing deep and the sun ascending, lean on familiar seasonal phrasing and add little originality to the sequence.
Granite Code 3B Instruct gap 3, disputed
Judged by J-O + J-A.
The response provides five stanzas rather than exactly three and does not follow 5-7-5 haiku syllable patterns. It also lacks a different colour word in each haiku and does not clearly make the third haiku anticipatory. The autumn subject is present, but several hard constraints are violated.
The response does maintain an autumnal subject and uses clear, grammatical language, with images involving leaves, crisp air, birds, and seasonal beauty. However, it substantially misses the requested form: it supplies five three-line stanzas, rather than three haiku, and the lines do not consistently match 5-7-5 syllables. Only one explicit colour word, “gold,” appears, and the final stanza offers a generic conclusion rather than anticipation. These are multiple direct constraint violations, despite the text remaining readable.
Asked for exactly three haiku in 5-7-5 form, each with a different colour word, third ending on anticipation. Output is five unstructured stanzas of free verse, not 5-7-5 haiku. Colour words are scarce (only 'gold' and 'bright'-type words, not clearly colour per stanza). Third stanza ends on 'perfect autumnal scene' and fifth gives a closing note, not anticipation. Fails the brief.
The response attempts a poem about late autumn and delivers stanzas with a seasonal theme, which shows some engagement with the subject. However, it produces five stanzas rather than three, and the lines do not follow a 5-7-5 syllable pattern. Colour words are weak: only gold appears as a clear colour, and the required distinct colour per haiku is absent. The final stanza closes on a sense of completion rather than anticipation, so the emotional requirement is missed. The language is fluent but generic, relying on stock images such as crisp air and bright sun.
Openthinker 7B V2 gap 3, disputed
Judged by J-O + J-A.
The response does not deliver the requested three haiku. It consists almost entirely of exposed planning and syllable-counting, ending with the incomplete fragment “Gold dust whispers on bare.” It therefore fails the output-only and completeness requirements, though it clearly attempts to address autumn, colors, syllable counts, and anticipation.
The response shows awareness of the core constraints: it considers distinct color words, checks syllable counts repeatedly, and aims for an anticipatory ending. However, the requested creative piece is never produced as a final answer. The output exposes internal drafting, includes many abandoned alternatives, violates the instruction to provide only three stanzas, and terminates mid-line. Because the response does not contain three completed haiku, its engagement remains an incomplete attempt rather than a usable submission.
The response is dominated by leaked deliberation and never delivers a clean set of three haiku. The visible haiku attempts break the 5-7-5 pattern, and the output is cut off mid-line in the third haiku, so no usable final text is provided.
The response consists mostly of visible planning notes rather than the requested piece. It does identify distinct colour words (crimson, amber, violet) and attempts a final haiku ending on hope, which reflects awareness of the brief. However, the haiku that appear do not meet the 5-7-5 syllable pattern, several lines are miscounted, and the text stops partway through the last stanza. The requested three-haiku sequence is not delivered as a clean final answer.
Qwen3.5-27B gap 3, disputed
Judged by J-O + J-A.
The response provides exactly three haiku in separate stanzas, with 5-7-5 syllable patterns throughout. Each stanza includes a colour word, and the third ends with anticipation of spring’s return rather than sadness. The imagery is clear and cohesive, though largely conventional for autumn poetry.
The response follows the requested structure precisely: it contains three untitled stanzas, each with apparently correct 5-7-5 syllable lines. Colour words are integrated naturally through “Crimson,” “Golden,” “Grey,” and “White,” while the final spring reference creates the requested anticipatory ending. Its main weakness is that much of the imagery is familiar, including drifting leaves, fading light, frost, and returning spring. The language remains clean and readable, but the sequence offers limited surprise or distinctive voice beyond its dependable seasonal progression.
All three haiku scan 5-7-5 on inspection, with colour words crimson, golden/grey and white. The third haiku closes on "Spring will soon return", meeting the anticipation requirement. Imagery is conventional (bare branches, frost creeping, winter sleeping beneath the ground), so the writing is competent but flat.
The response delivers three correctly structured haiku, each with a colour word, and the final stanza ends on anticipation of spring rather than sadness, so every hard constraint is met. Its strength lies in this clean compliance and clear seasonal progression from crimson leaves to frost to snow. Its weakness is the stock imagery: phrases such as winter sleeping beneath the ground and spring returning are familiar tropes, and the second haiku uses two colour words, so the verses lack a fresh observation or distinctive voice.
Smollm2 360M Instruct gap 3, disputed
Judged by J-O + J-A.
The response attempts autumn-themed haiku but does not deliver exactly three stanzas. It contains a long sequence of repetitive, mostly non-haiku stanzas, fails the 5-7-5 syllable structure, does not provide three distinct colour words in three haiku, and ends abruptly mid-stanza.
The response does engage with the requested autumn subject and repeatedly uses nature imagery such as wind, leaves, sun, and moon. However, it substantially fails the format: it supplies many repeated stanzas instead of exactly three, several lines do not follow 5-7-5 syllables, and the output is truncated at the end. The colour-word and anticipation requirements are also not coherently satisfied across three completed haiku. Its repetition makes the sequence feel mechanically expanded rather than intentionally composed.
The output begins with three stanzas that are not haiku (syllable counts far off 5-7-5 and no clear anticipation ending), then degenerates into hundreds of repeated stanzas and ends mid-line with 'The autumn'. No usable, correctly formed haiku sequence is delivered, and the output is cut off.
The response opens with three stanzas, but they do not follow the requested 5-7-5 syllable pattern, and the third does not close on anticipation. The stanzas then repeat the same few lines over and over, with no new haiku emerging, and the text breaks off mid-phrase. The colour requirement is only loosely touched: golden and green appear, but the stanzas lack any distinct structure. The main weakness is that the requested exactly-three-haiku deliverable is never produced cleanly, and the output is degenerate and truncated.
Exaone Deep 2.4B gap 2
Judged by J-O + J-A.
The response discusses the requested haiku and proposes individual lines, but it never supplies exactly three finished stanzas. It remains an extended, repetitive planning monologue, ends mid-thought, and violates the required output-only format.
The response shows awareness of the core constraints, including the 5-7-5 pattern, distinct colour words, and the third haiku’s anticipatory tone. However, it never converts that planning into the requested final sequence. Instead, it repeats alternatives, miscounts several syllables, introduces multiple colours within proposed stanzas, and ends abruptly while still deliberating. The required three haiku are therefore absent, and the output contains extensive internal reasoning rather than only the requested poem.
The output is dominated by visible reasoning that loops without settling on a final answer. The model never delivers three clean haiku as a final response; it ends mid-sentence while still working through syllable counts. No usable haiku sequence is delivered.
The response consists of an extended, repetitive planning monologue rather than the requested output. It considers colour assignments and syllable counts at length, and some fragments of candidate verses appear, but none is presented as a finished three-stanza sequence. The text ends mid-sentence without any final haiku. The one aspect of the brief engaged with is the colour requirement, where it correctly identifies that each haiku needs its own distinct colour word. Nonetheless, the requested piece is not delivered, and the output is truncated and includes leaked reasoning.
Gemma 2B Instruct V1.1 gap 2
Judged by J-O + J-A.
The response provides three stanzas, but it violates the text-only requirement with a title. The second haiku does not follow 5-7-5 (the second and third lines are short), and the second line of the third haiku is also short. The ending suggests approaching winter rather than clear anticipation.
The response’s strength is that it supplies three recognizable haiku stanzas centered on late autumn, with distinct colour terms including crimson, golden, and scarlet. Its weaknesses are several hard-constraint violations: an unrequested title appears, the second and third haiku do not consistently meet the required syllable pattern, and the final image emphasizes winter’s chill rather than anticipation. The imagery is understandable but largely conventional, relying on familiar autumn vocabulary and broad statements instead of a more specific, distinctive turn toward what comes next.
Response includes an unrequested title despite the brief asking for no title. Haiku syllable counts are off: line one of the first haiku has 4 syllables, and the second haiku's middle line is short. Colour words are present (crimson, golden, scarlet) but the third haiku ends on winter's chill, a sad, not anticipatory, note. Hard constraints violated repeatedly.
The response attempts three haiku on autumn with a distinct colour word in each, which is the basic shape requested. However, it adds a title the brief explicitly forbade, and several lines miss the 5-7-5 syllable pattern, such as the opening line of the first stanza. The final haiku closes on winter's chill and near-end imagery, which reads as melancholy rather than the anticipation the brief required. Imagery is generic, drawing on stock phrases like vibrant grace and nature's masterpiece, with little specific observation of the season's last week.
Gemma3 1B IT gap 2
Judged by J-O + J-A.
The response provides exactly three haiku in separate stanzas, with 5-7-5 syllable lines, distinct colour vocabulary, and no title. The final haiku ends with anticipation through “Future’s coming soon.” The imagery is clear and seasonal, though some phrasing is familiar and the first haiku uses two colour words rather than one.
The response fulfills the formal structure effectively: each stanza follows a recognizable 5-7-5 pattern, the colour words vary across the sequence, and the final line shifts toward expectation rather than mourning. Its seasonal images, including descending leaves, lengthening shadows, and returning sunlight, are coherent and easy to follow. A weakness is that several expressions are conventional, such as leaves descending and promises in dusk, which limits the sense of distinctive poetic discovery. The first stanza also includes both “Golden” and “Crimson,” making its colour assignment less sharply differentiated than the later stanzas.
All three haiku scan 5-7-5 on inspection, each uses a different colour word (golden, russet, amber, plus crimson), and no title is given. The third closes on "Future's coming soon", meeting the anticipation requirement. The imagery is stock (golden leaves, promise, sunlight), so the writing is competent but flat.
The response meets every hard constraint: three stanzas, correct syllable counts, distinct colour words, and an anticipatory close in the third haiku. Its strength is this structural accuracy, and the second and third stanzas move from fading light toward a sense of something to come. Its weakness is that the imagery is largely expected. Golden leaves, crimson whispers and a promise in the dusk are familiar autumn tropes, and the third haiku's ending, "Future's coming soon", states its anticipation without a specific image to carry it.
Granite Code 8B Instruct gap 2
Judged by J-O + J-A.
The response provides three autumn-themed stanzas and uses different colour terms, but it does not deliver three haiku: the second and third stanzas have four lines, several lines miss the 5-7-5 syllable pattern, and the third ends with nostalgia rather than clear anticipation.
The response does engage with the requested autumn setting and includes multiple distinct colour words, such as orange, goldenrod, and crimson. However, it violates the central formal constraint: only the first stanza resembles a three-line haiku, while the other two contain four lines and inconsistent syllable counts. The final stanza also closes on nostalgia for cozy comforts, which is more reflective than anticipatory. The three-stanza structure is present, but the requested exact haiku sequence and ending mood are not consistently achieved.
Syllable counts are violated throughout (e.g. 'Orange leaves falling' is 5, 'Autumn's last week' is 5 but the second haiku has 3 lines of varying length and a fourth line). The brief asked for three 5-7-5 haiku, but the output has three stanzas with 3, 4 and 4 lines. Colour words are present, but cherry blossoms are spring imagery and the third stanza ends on nostalgia, not anticipation.
The response attempts three stanzas and uses a colour word in each, which partly engages with the brief. However, the stanzas do not follow the required 5-7-5 structure, with the second and third running to four lines each and mismatched line lengths. The third stanza closes on nostalgia rather than anticipation, contrary to an explicit requirement. Imagery such as cherry blossoms also conflicts with the autumn setting. The writing is clear and readable, but the core constraints are largely unmet.
granite-4.1-3b gap 2
Judged by J-O + J-A.
The response provides only two stanzas instead of exactly three. It also misses the required 5-7-5 pattern in lines such as “crimson hills whisper night” and “wistful shadows stretch long.” The existing imagery suits late autumn and includes distinct colour words, but the required third haiku and its anticipatory ending are absent.
The response uses concise autumn imagery, including falling golden leaves, crimson hills, and fading amber light, and its two stanzas are generally coherent. However, it does not deliver the requested sequence of exactly three haiku: the entire third stanza is missing. Several lines also fail the specified syllable pattern, particularly the second lines of both surviving stanzas. Because the missing stanza would have needed to end with anticipation, that thematic requirement cannot be satisfied or assessed from the submitted text.
Only two stanzas are delivered, not three. Line syllable counts are off (haiku 1 reads about 5-6-4, haiku 2 about 5-6-5). Colour words differ (golden, crimson, amber). The final line, 'winter's breath lingers', closes on wistfulness rather than anticipation.
The response offers two stanzas of haiku instead of the three requested. Its strength is that the colour words golden, crimson and amber are distinct across the stanzas, and the imagery of autumn leaves and fading light is recognisable. Its main weakness is the structure: the line lengths drift from the 5-7-5 pattern, with the middle line of the first stanza and the final line of the second both short or long. The closing image, winter's breath lingering, reads as melancholy, so the requested note of anticipation is not achieved.
Granite4 1B H gap 2
Judged by J-O + J-A.
The response supplies only two lines of one incomplete haiku rather than exactly three complete haiku. The first line has four syllables, and the required colour variation and anticipatory third stanza are absent. It is a partial attempt on the autumn subject but does not deliver the requested sequence.
The response does establish an autumnal image through golden leaves and a crisp breeze, and its second line appears to follow the requested seven-syllable pattern. However, it stops after two lines, leaving the first haiku incomplete and omitting the other two haiku entirely. Consequently, it does not provide the required three stanzas, the distinct colour words, or the anticipatory ending. The output also fails the requested complete-form constraint because the first line contains four syllables rather than five.
Only two lines are delivered, not three haiku. The response contains a single incomplete fragment with one colour word (golden), no third haiku, no 5-7-5 structure completed, and no anticipation ending. The requested piece is substantially missing.
The response attempts the requested form with an opening line that evokes autumn imagery and includes a colour word, and the two lines read fluently. However, it stops after two lines, so only a fragment of the first haiku exists. The second and third haiku are absent, the 5-7-5 structure cannot be verified as complete, and the required anticipation ending is never reached. The piece is therefore not delivered, and the response is cut off well before its end.
Granite4 350M H gap 2
Judged by J-O + J-A.
The response supplies three stanzas, but the lines do not consistently follow 5-7-5 syllable patterns. The first stanza uses two colour words, while the second and third use none, so the colour requirement is unmet. The third stanza ends in peaceful closure rather than anticipation.
The response does provide three recognizable three-line haiku-like stanzas and uses autumn imagery in the first and third, showing engagement with the requested subject. However, it misses several explicit constraints: syllable counts are substantially inaccurate, colour words are not distributed one per haiku, and the second stanza shifts to cherry blossoms, suggesting spring. The final stanza concludes with “A peaceful end,” which conveys closure rather than anticipation. The piece is readable, but its structural and thematic compliance is limited.
Three stanzas are delivered, but syllable counts miss 5-7-5 (e.g. 'Nature's beauty.' is 4 syllables, 'Cherry blossoms dance' is 6). Colour words appear only in stanza one ('Crimson and gold'), not in each haiku. Stanza two's 'Cherry blossoms' is a spring image, and the third ends on 'A peaceful end', which is closure rather than anticipation.
The response supplies three stanzas, which matches the requested structure, and the opening image of crimson and gold in the first stanza is a recognisable autumn choice. However, the stanzas do not follow the 5-7-5 syllable pattern, and only the first contains a colour word, so the brief's constraint that each haiku include a different colour is not met. The second stanza's cherry blossoms suggest spring rather than the last week of autumn. The final stanza closes on a peaceful end rather than anticipation, which fails the ending requirement and leaves the sequence's arc unresolved.
Internlm2 1.8B Chat V2.5 gap 2
Judged by J-O + J-A.
The response supplies only one stanza rather than exactly three haiku. Its lines also do not consistently meet 5-7-5 syllable counts, repeat “golden” instead of using three different colour words, and provide no third-haiku anticipation. It is nevertheless recognizably about autumn and formatted as a titleless stanza.
The response makes a recognizable attempt at an autumn haiku and maintains a compact, image-based form, with leaves, sky, and light providing a coherent seasonal scene. However, it delivers only one stanza instead of the required sequence of three, and its first two lines appear to exceed the requested syllable counts. The repeated use of “golden” also fails the requirement for different colour words, while the missing third haiku means there is no concluding note of anticipation.
Delivers one three-line unit, not three haiku. Lines are not 5-7-5 syllables, the same colour word 'golden' repeats in every line rather than a different colour per haiku, and there is no anticipatory third haiku. Only fragmentary attempt at the requested form is visible.
The response attempts a short poem about autumn and keeps a consistent image of golden light, which is the only element that partly engages the brief. However, it supplies a single three-line stanza instead of three separate haiku, and its lines do not follow the 5-7-5 syllable pattern. The colour word 'golden' is repeated rather than varied across the sequence, and the closing line offers no sense of anticipation. The piece therefore does not deliver the requested sequence of three haiku, though it is readable and coherent.
Mathstral 7B V0.1 gap 2
Judged by J-O + J-A.
The response provides three autumn-themed stanzas, but none consistently follows 5-7-5: for example, the first lines scan as 7, 7, and 5 syllables. The third haiku lacks a colour word and ends in winter's approach rather than anticipation. Numbering also conflicts with the instruction to output only the haiku.
The response does establish an autumn setting and uses distinct colour terms in its first two stanzas, including red, orange, and golden imagery. However, the central formal requirement is not met: the lines regularly exceed the requested syllable counts, and the third stanza contains no colour word. Its ending emphasizes winter’s arrival and a final farewell, which reads as somber rather than anticipatory. The numbered presentation is also inconsistent with the request for only three haiku, one per stanza, without additional formatting.
The response is numbered and uses lines that do not follow 5-7-5 syllables (e.g. 'Nature's art, so grand.' is about 5 syllables but the second line is longer; third haiku 'As the days grow short' is 6 syllables). Colour words appear (red, orange, golden) but the third haiku ends on 'Winter's approach, a sort', which is incoherent and not an anticipatory note. Numbering violates the no-title/plain-output spirit only mildly, but the meter and closing requirement fail.
The response attempts three numbered verse stanzas on autumn and introduces a different colour word in each: red, orange and golden. That partly meets the brief's structure. However, the syllable pattern is not held: lines vary in length and do not follow 5-7-5 in several places. The third stanza ends on a closing line about winter approaching, which reads as a vague, somewhat sombre note rather than anticipation, and the phrasing 'a sort' is unclear. The verse also relies on stock imagery and loose rhyme, and the wording is uneven throughout.
Muse Glimmer 30B gap 2
Judged by J-O + J-A.
The response supplies three thematically appropriate haiku with distinct colour words—gold, crimson, and white—and the final poem anticipates winter. However, the extra opening line, “Last week of autumn,” functions as an unrequested title or fourth line, violating the instruction to output only three haiku with no title.
The response has a clear seasonal progression and successfully uses a different colour word in each haiku. Its strongest structural choice is the final image of winter waiting to wake, which provides anticipation rather than sadness. The main weakness is the added opening line, which acts as a title and means the output is not only the three requested stanzas. The individual haiku otherwise follow the requested 5-7-5 pattern and remain readable, though some imagery, such as bleeding maples and early night, is familiar rather than especially distinctive.
All three haiku scan 5-7-5 on inspection. Colour words gold, crimson and white are distinct. The third haiku closes on 'winter waits to wake', an anticipatory note. No title, and the stanzas are separated. Imagery such as 'crows count the cold days' is competent but largely expected.
The response meets the structural constraints: three stanzas, correct syllable counts, a different colour word in each, and no title. The third haiku ends on anticipation, as the brief asks. Its strength is clarity and a clean seasonal progression from gold leaves to frost and first stars. Its weakness is that much of the imagery is conventional, with the opening line 'Last week of autumn' stating the subject rather than rendering it. The second haiku's closing line, 'night comes early now', also leans on a familiar autumn image.
Nemotron Mini 4B Instruct gap 2
Judged by J-O + J-A.
The response provides only one three-line stanza rather than exactly three haiku. It includes colour words, but not three different colours across three haiku, and its first line appears to have four syllables rather than five. The autumn subject is recognizable, but the required third-haiku anticipation is absent.
The response does establish the requested autumn setting and uses a clear, seasonal image of golden and red leaves. Its main weakness is substantial noncompliance with the form: it supplies one stanza instead of three haiku, lacks the required progression toward anticipation, and does not demonstrate three distinct colour words assigned across separate haiku. The opening line also appears metrically short, so even the submitted stanza does not consistently meet the 5-7-5 requirement.
Only one three-line poem is delivered, not the required three haiku. Syllable counts are not 5-7-5 (e.g. 'Autumn's last week' is 5 but 'Golden leaves dance in the breeze' is 8). Colour words appear (Golden, Red) but only two haiku exist, and the third anticipation requirement is unmet. Writing is readable but the brief is substantially failed.
The response delivers a single three-line verse rather than the requested sequence of three haiku, so the core structure of the brief is missing. The one identifiable strength is that it stays on the subject of late autumn and uses colour imagery, with golden and red leaves appearing in the lines. However, the syllable pattern does not follow 5-7-5 in its lines, and no third haiku ends on a note of anticipation. The text is fluent enough to read easily, but it does not meet the constraints set by the brief.
Olmo-3-7B-Think gap 2
Judged by J-O + J-A.
The response provides exactly three haiku, uses three distinct colour words, and maintains the requested autumn subject and anticipatory ending. The first two haiku follow 5-7-5 syllable patterns, but the final line, “Spring's first bloom awaits,” has six syllables rather than five.
The response’s main strength is its clear fulfillment of the structural and thematic brief: it presents three separate stanzas, each with a distinct colour word, and the final haiku turns toward spring rather than ending in grief. Its imagery is coherent, particularly the skeletal branches and approaching harvest. The decisive weakness is a syllable-count error in the third haiku’s final line, which scans as six syllables. Because the task requires every haiku to be 5-7-5, this single constraint violation prevents the response from fully satisfying the brief.
All three haiku follow 5-7-5 syllables when counted. Colour words differ (crimson, golden, amber). The third ends on 'awaits', meeting the anticipation requirement. No title is given. Weaknesses include the repeated 'whisper/whispers' and a muddled 'hope's fading breath' line.
The response meets the hard constraints: three stanzas, correct syllable counts, a different colour word in each, and a closing line that turns toward anticipation with 'Spring's first bloom awaits'. The image 'Skeletal branches hold moon' is the most distinctive moment, giving a concrete, unexpected picture. The writing is weaker elsewhere. 'Crimson leaves whisper' and 'Whispers of the falling breeze' repeat the same word within the first two lines, and the middle line of the third haiku, 'Leaves cling to hope's fading breath', is muddled and leans on stock sentiment.
Phi3 14B Medium 128K Instruct gap 2
Judged by J-O + J-A.
The response provides exactly three haiku in separate stanzas, with consistent 5-7-5 syllable patterns. Each uses a distinct colour word—“Golden,” “Crimson,” and “White”—and the final haiku ends with anticipation through spring’s promise. The imagery is clear and appropriate, though familiar and somewhat conventional.
The response fulfills the structural requirements cleanly: it presents three separate haiku, uses three different colour words, and maintains the requested syllabic pattern. The third stanza shifts toward anticipation with the arrival of snow and the promise of spring rather than ending in grief. Its main weakness is originality, since phrases such as fading sunsets, barren trees, quiet sighs, and spring’s promise are familiar seasonal images. The language remains readable and tonally consistent, but the sequence does not develop a particularly distinctive voice or surprising perspective on autumn’s final week.
Three stanzas with a colour word each (golden, crimson, white) and an anticipatory close ('Spring's promise awaits'). The third haiku's middle line, 'White blankets the slumbering earth', runs about eight syllables, breaking the 5-7-5 constraint. The first two stanzas scan correctly.
The response delivers three stanzas in the requested form, each with a distinct colour word, and the first two follow the 5-7-5 pattern closely. Its strongest moment is the closing line of the third haiku, which turns toward spring rather than loss, as the brief asks. The main weakness is the third haiku's middle line, which has roughly eight syllables instead of seven. The third stanza also shifts to snow and spring imagery, which sits awkwardly with the last week of autumn.
Qwen 0.5B Chat V1.5 gap 2
Judged by J-O + J-A.
The response provides only one three-line fragment rather than exactly three haiku. Its first line has four syllables, the second has five, and the Chinese final line does not establish the required 5-7-5 English haiku structure. It includes one colour word and suggests anticipation, but omits the other required haiku and distinct colours.
The response does attempt the requested autumn subject, using red leaves and a farewell image, while the final Chinese line appears to gesture toward anticipation of a new season. However, it delivers only one stanza instead of three, and the syllable pattern is not met even in the English lines. The mixed-language final line also makes the haiku’s meter and meaning difficult to verify. Consequently, the central structural requirements remain unfulfilled despite the recognizable thematic attempt.
The response contains only two English lines and one Chinese line, which is not three haiku. The lines do not follow a 5-7-5 pattern, the stanza structure is absent, and the Chinese line ('anticipating the arrival of the new season') is off-brief and untranslated. Only one colour word appears, and no third haiku ends on anticipation in a proper form.
The response attempts to address the autumn theme through the image of red leaves falling, which is a recognisable seasonal cue. However, it fails the core task: it does not deliver three haiku, the lines do not follow the required 5-7-5 syllable structure, and only one colour word appears where each haiku needed a different one. The closing line switches into Chinese without explanation, breaking clarity for an English-language brief. The piece is incomplete and cannot be counted as a usable sequence.
Qwen2 0.5B Instruct gap 2
Judged by J-O + J-A.
The response provides only one stanza instead of exactly three haiku. Its first line has five syllables, but the second has about five rather than seven and the third about seven rather than five. It includes multiple colour words in one haiku, not different colours across three haiku, though its final phrase gestures toward anticipation.
The response does attempt the requested haiku form and evokes autumn through falling leaves, while its closing phrase suggests anticipation rather than sadness. However, it does not deliver the required sequence: two haiku are missing, and the syllable pattern is incorrect in the second and third lines. The colour requirement is also unmet because several colours appear together instead of one different colour in each of three haiku. The output therefore engages with the subject but fails the central structural constraints.
Delivers only one three-line stanza instead of three haiku. The line counts are roughly 4-5-5 syllables, not 5-7-5, and the lines are not a valid haiku structure. Colour words appear (red, orange, yellow) but only in one stanza, and the anticipation note is present but in a single unit, so the requested sequence is not delivered.
The response offers a single three-line verse rather than the requested sequence of three haiku. Its lines do not follow the 5-7-5 syllable pattern; the second line lists three colours, which is a faithful nod to the colour requirement, but it is applied within one stanza instead of across three separate haiku. The final line gestures toward anticipation, which fits the brief's closing note, yet the overall piece is substantially short of the three-stanza structure the task specified, so the core deliverable is missing.
Qwen2 Math 1.5B Instruct gap 2
Judged by J-O + J-A.
The response recognizes the autumn-and-colour premise and presents three identifiable haiku attempts, but it violates the output-only requirement with extensive commentary and malformed LaTeX, includes many additional haiku beyond three, and does not make the third haiku anticipatory. Several lines also fail the required 5-7-5 syllable pattern.
The response’s main strength is that it identifies the requested structure and supplies colour words in distinct attempted stanzas. However, it does not deliver a usable final sequence: it includes planning commentary, numbered explanations, a purported final answer, and numerous extra embedded haiku. The presented examples also contain syllable-count errors, and the third haiku focuses on blue objects rather than anticipation. Because the response is a recognizable but substantially noncompliant attempt, it only partially engages with the brief.
The response opens with preamble and numbered explanation, then gives haiku that are not 5-7-5 (e.g. 'Red wine in the glass' is 5 but 'Red flowers in the sky' is 5, 'Blue sky in the sky' is 5 with dubious lines), repeats the same stanzas, uses no anticipatory ending, and wraps the output in malformed LaTeX with a fourth-and-fifth set of extra haiku. Constraints of exactly three haiku and no extra text are violated.
The response attempts a haiku sequence and does name colour words for each piece, which is the only real engagement with the brief. However, it fails it substantially: it adds an explanatory preamble and numbered plan despite the instruction to output only the haiku, and the final block contains malformed LaTeX with more than three haiku, including repeated and duplicated stanzas. The third haiku about blue skies does not end on anticipation, and the verses contain no seasonal imagery of late autumn beyond bare colour lists. Several lines are repetitive, with phrases like 'in the sky' recurring, so the sequence reads as a mechanical template rather than crafted verse.
Qwen2.5 Coder 0.5B Instruct gap 2
Judged by J-O + J-A.
The response provides only one three-line stanza rather than exactly three haiku. Its lines do not consistently follow 5-7-5 syllables, include no colour word, and omit the required anticipatory ending. It is nevertheless recognisably an attempted autumn haiku.
The response makes a recognizable attempt at an autumn-themed haiku and uses concise seasonal imagery, particularly leaves and silence. However, it supplies only one stanza instead of three, and the lines do not meet the required 5-7-5 syllable pattern. It also lacks the required different colour words and does not provide a third haiku ending with anticipation. These omissions prevent it from fulfilling the central structural and thematic constraints of the brief.
The response offers a single three-line fragment rather than three haiku. It does not follow 5-7-5 syllable structure, contains no colour word, and does not end on anticipation. Only the first line is loosely on-topic, so the requested piece is not delivered.
The response attempts a short autumnal verse, and its imagery of falling leaves and whispering silence gestures toward the requested subject. However, it supplies one three-line stanza instead of the three required haiku, so the requested sequence is not delivered. None of the lines follows a 5-7-5 syllable pattern, no colour word appears, and the closing line carries a quiet, unresolved tone rather than the anticipation the brief asks for. The text is readable, but it fails the brief on structure and on several hard constraints.
Qwen2.5 Coder 1.5B Instruct gap 2
Judged by J-O + J-A.
The response offers only three lines rather than exactly three haiku in separate stanzas. It includes one colour word, “Golden,” but lacks two additional haiku and therefore fails the distinct-colour and third-haiku anticipation requirements. The first line has five syllables, but the remaining lines do not form a 5-7-5 haiku.
The response makes a recognizable attempt at autumnal haiku imagery, particularly through “Golden leaves” and the quiet atmosphere of whispering. However, it supplies only one incomplete stanza instead of three haiku, and its second and third lines do not meet the required syllable counts. It also provides only one colour word, omits the required additional colours, and cannot establish an anticipatory ending in a third haiku. The output therefore engages with the subject but does not deliver the requested sequence or satisfy its structural constraints.
Only one three-line poem is delivered, not three haiku. The lines do not follow 5-7-5 syllables, the colour word 'Golden' appears once and the others have none, and the 'whispers' repetition weakens the piece. The third haiku is absent, and no anticipation note is present.
The response attempts a short autumnal verse and uses the colour word 'Golden' in its opening line, which is the only element that partly engages with the brief. However, it delivers a single three-line stanza rather than the required sequence of three haiku. The lines do not meet the 5-7-5 syllable pattern, each haiku needs a distinct colour word, and the repeated 'whispers' makes the imagery flat. No closing note of anticipation is given, so the piece fails the brief's central requirements.
Qwen3 0.6B gap 2
Judged by J-O + J-A.
The response provides three autumn-related poetic lines, but they are not three haiku: the lines do not follow 5-7-5 syllable patterns, and the numbered single-line format omits the required three-line stanzas. The colour terms are also not clearly three different colours, and the third line's anticipation is underdeveloped.
The response does engage with autumn imagery and includes colour-related language such as “Golden” and “Silver,” while its final line gestures toward the future through the next dawn. However, it does not deliver the requested form: each item is only one line rather than a 5-7-5 haiku, and numbering conflicts with the instruction to output only three stanzas. The colour words are not clearly three distinct colours, and the last line is too generic to establish a strong anticipatory note. The piece is therefore recognisable as an attempt but substantially incomplete.
Three numbered lines rather than three stanzas of haiku. Syllable counts miss 5-7-5: line 1 is about 12 syllables, line 2 about 11, line 3 about 8. Colour words appear (golden, silver, gold) but golden and gold repeat across haiku, breaking the different-colour rule. Third line ends on 'dawn' with a hint of anticipation, but the form is wrong throughout.
The response offers three numbered lines that loosely gesture at autumn imagery and a closing hint of anticipation in the final line, which is the one element that partly matches the brief. However, it does not produce haiku: each line is a full sentence rather than the 5-7-5 structure, and the lines are far longer than the required syllable counts. The colour requirement is also undermined, since 'golden' and 'gold' appear across the first two haiku, so the colour words are not distinct. The text is clear and readable, but the requested form is not delivered.
Sailor2 8B Chat gap 2
Judged by J-O + J-A.
The response supplies three autumn-themed stanzas and the third gestures toward anticipation. However, it violates the text-only requirement with titles and a parenthetical, and the lines do not consistently follow 5-7-5 syllable counts. The third stanza also uses multiple colour words rather than one distinct colour per haiku.
The response has a recognizable autumn progression, with amber, red, violet, and emerald imagery creating a seasonal palette. Its final stanza also shifts toward spring and light, which supports the requested anticipatory ending. However, the required haiku form is not maintained: several lines clearly exceed or fall short of the 5-7-5 pattern. The added titles and the explicit “(Anticipation)” note also conflict with the instruction to output only three untitled haiku stanzas.
Brief requires exactly three haiku, no title. Response adds three bold titles and an annotation '(Anticipation)'. Syllable counts are off in several lines (e.g. 'Sunsets paint the sky's gold' is 7 syllables, 'Whispers of harvest, leaves' farewell' is 9). Colour words present but the third haiku's ending is an unrequested winter image rather than clearly anticipatory.
The response supplies three stanzas, each using a different colour word (golden, crimson, emerald, amber, red, violet), which partially meets the core structure. However, it violates the no-title instruction by adding three bold headings and an unrequested parenthetical annotation. Several lines do not follow the 5-7-5 syllable pattern, such as the closing line of the first haiku and the middle line of the first and third poems. The third haiku's closing image of winter's door is only loosely anticipatory and reads as a conventional phrasing. Overall the piece is recognisable as haiku but fails several explicit constraints.
Smallthinker 3B Preview gap 2
Judged by J-O + J-A.
The response supplies three recognizable autumn haiku, uses distinct colour words, and makes the third anticipate spring. However, it violates the required output-only format with extensive planning text and a duplicated final answer. Several lines also miss 5-7-5, including “Autumn's last breath” (4), “Orange sunsets” (4), and “Paint the sky with warm hues” (6).
The response does engage directly with the requested subject and creates a clear seasonal progression from fading autumn toward spring. Its strongest element is the third stanza, which uses “Green” and explicitly shifts toward anticipation through emerging shoots and spring’s promise. The main weakness is constraint handling: the answer includes planning commentary, a “Final Answer” label, and repeats all three stanzas, contrary to the instruction to output only three haiku. In addition, the first two haiku contain lines that do not follow the required 5-7-5 syllable pattern.
The output opens with a long visible deliberation the brief did not ask for, then repeats the haiku in a 'Final Answer' block, violating the text-only instruction. The haiku syllable counts are off (e.g. 'Autumn's last breath' is roughly 5, but 'Red leaves crunch beneath my feet' is 7, while 'Golden days remain' is 5; 'Orange sunsets' lines miscount). The third haiku does not clearly end on anticipation.
The response attempts three haiku with a distinct colour in each (red, orange, green) and a recognisable seasonal subject. Its main weakness is that it prints extended planning notes before the poems and then repeats the verses under a 'Final Answer' heading, despite the brief asking for only the three haiku with no extra text. Several lines also miscount syllables against the 5-7-5 requirement, and the closing haiku's anticipation is weakly conveyed. The stanza structure is otherwise clear.
The other 198 models
Their judges were closer together. Their scores are listed here; open the model's page for the full answer.
Withheld
These models are kept out of the ranking, so none of their scores or reasons are shown:
The judge room names the judges and shows where they disagree most.