Skip to main content

← Dracon Bench leaderboard

dots-studio/dots-3-note-preview:free

Tested First tested 2026-11-09 · OpenRouter free lane

Dots Studio's Dots 3 note preview carries the best GSM8K-50 in the spike at 92.0% with zero failures. No published IFEval or GSM8K reference found, so these are novel numbers. SVG total 126/250, split between the run's best clock (43/50) and its weakest monalisa (12/50).

Scores

SuiteScore
IFEval-50 prompt-strict86.1% (n=36)
IFEval-50 instruction-strict88.5% (n=36)
GSM8K-5092.0% (46/50)
SVG blind rubric126/250

Artifacts

Blind-judged by nvidia/nemotron-3.5-super-vl:free: the judge saw the brief and the rendered image only, never the model name. Briefs are frozen; expand one to read it.

beach · 24/50

Brief fidelity (4/10)
Most requested elements are present (sun, reflection, palm, sailboat, birds, starfish, sand), but the two dolphins arcing from waves are missing, the black ovals do not read as dolphins, so the missing-headline-element cap applies.
Construction (5/10)
Shapes are simple silhouettes and primitives rather than assembled anatomy; the palm and boat read clearly, but the dolphin shapes are just featureless ovals and nothing appears melted or duplicated.
Composition (5/10)
The layout is balanced with the sun centered, palm left, and boat right, but perspective is flat with no atmospheric haze or size-based depth for distant objects.
Detail finish (4/10)
The sky gradient and solar glow are pleasant, but the water is flat teal, the reflection is a simple vertical gradient without shimmer texture, and small touches are minimal.
Visual appeal (6/10)
The warm orange-pink-to-teal palette is coherent and the minimalist style has some charm, but the overall image is quite basic and lacks confident detail.
Brief

Draw a beautiful sunset beach scene: a large sun sitting low over the sea with a shimmering reflection path on the water, a leaning palm tree silhouette on the left, two dolphins arcing out of the waves, a small sailboat on the right, birds in the sky, sandy foreground with a starfish. Warm oranges and pinks fading to teal. Coherent perspective: distant objects smaller and hazier than near ones. Output: exactly one self-contained SVG (viewBox "0 0 800 600", no external references) inside a single ```svg fenced block. Short prose outside the fence is allowed.

clock · 43/50

12 1 2 3 4 5 6 7 8 9 10 11 DRACON
Brief fidelity (9/10)
All requested elements are present: 12 ordered numerals, minute tick marks, hour hand near 10, minute hand at 2, red second hand at 6, and DRACON text under the center, though the hour hand is not distinctly 'just past' 10.
Construction (9/10)
The clock face, hands, and numerals are cleanly assembled vector shapes with no melting, duplication, or structural errors.
Composition (9/10)
The clock is perfectly centered within the 800x600 frame with balanced proportions between the dark rim, light face, and internal elements.
Detail finish (8/10)
Flat vector execution includes a dotted minute track, contrasting red second hand, and legible DRACON lettering with no broken edges or stray shapes.
Visual appeal (8/10)
The minimal black, off-white, and red palette is coherent and confident, producing a clean, functional, and charming clock illustration.
Brief

Draw a round wall clock face showing exactly 10:10: a circular dial with all 12 hour numerals in the correct positions, minute tick marks, an hour hand pointing just past 10, a minute hand pointing exactly at 2, and a thin second hand pointing at 6. The word DRACON in small capitals under the center. Dark rim, light face, flat vector style, centered composition. Output: exactly one self-contained SVG (viewBox "0 0 800 600", no external references) inside a single ```svg fenced block. Short prose outside the fence is allowed.

duo · 16/50

Brief fidelity (3/10)
The image includes sun, clouds, water, a horizon line, and two figures on an orange craft, but the dog, raccoon, paws on handlebars, and leaning turn are unrecognizable or absent, so headline elements are missing.
Construction (2/10)
The characters are featureless brown blobs without faces, paws, or distinct anatomy, and the jet ski is a simple orange oval with no handlebars, so nothing looks assembled.
Composition (4/10)
The horizon divides the frame cleanly and the sun balances the upper right, but the subjects are small, flat, and lack clear perspective or dynamic placement.
Detail finish (3/10)
There is minimal shading or texture, no visible handlebars or facial features, and the sea spray is just a row of plain white circles with no broken markup evident.
Visual appeal (4/10)
The bright cartoon palette is cheerful, but the crude, unrecognizable figures and simplistic shapes reduce charm and confident linework.
Brief

Draw a comical scene: a dog driving a jet ski at speed, paws on the handlebars, with a raccoon riding behind holding on, both leaning into a turn. Sea spray behind the jet ski, sun and clouds in the sky, a distant shoreline. Both characters must be clearly doing something, not just posed side by side. Bright cartoon vector style. Output: exactly one self-contained SVG (viewBox "0 0 800 600", no external references) inside a single ```svg fenced block. Short prose outside the fence is allowed.

monalisa · 12/50

Brief fidelity (3/10)
The figure, dark dress with neckline, and smile are present, but the view is frontal rather than three-quarter, the hands are an indistinct blob, and the background lacks a clear winding river and blue mountains.
Construction (2/10)
The head, neck, torso, and hands are assembled from simple geometric primitives, but anatomy is crude with no shoulders or arms and the hands are unrecognizable.
Composition (3/10)
The portrait is centered and fills the frame, but the head is oversized relative to the body and the background layers lack convincing perspective.
Detail finish (2/10)
The artwork is flat with no shading, textures, or refined small touches, and the bottom element appears cut off by the frame edge.
Visual appeal (2/10)
The muted palette is coherent, but the execution is so primitive and schematic that it lacks charm or confident linework.
Brief

Redraw the Mona Lisa as a flat vector portrait, bust composition: a seated woman in three-quarter view facing the viewer, dark dress with neckline detail, hands folded at the bottom edge, soft enigmatic smile, hair parted in the middle. Behind her a hazy landscape with a winding river, a bridge on the right, and blue mountains. Muted greens, browns, and smoky blues. Output: exactly one self-contained SVG (viewBox "0 0 800 600", no external references) inside a single ```svg fenced block. Short prose outside the fence is allowed.

pelican · 31/50

Brief fidelity (4/10)
The pelican, profile pose, full body, neck, wing, and feet are present, but the throat pouch and sea horizon are missing headline elements, capping the score at 4.
Construction (7/10)
The bird is built from clean geometric shapes that look assembled rather than melted, with no duplicated or broken parts, though the anatomy is highly abstracted.
Composition (7/10)
The pelican is well-placed in the frame facing left with a clear ground line and sky, using the space effectively for a flat side-profile illustration.
Detail finish (6/10)
The artwork is clean with no stray shapes or cut-off edges, but it lacks the throat pouch, sea horizon, and any shading or textural details from the brief.
Visual appeal (7/10)
The limited blue-brown-orange palette is coherent and the confident flat shapes give the piece simple charm.
Brief

Draw a brown pelican standing in side profile, facing left, full body visible: large beak with throat pouch, long neck, folded wing, webbed feet on flat ground. Simple sky background with a hint of sea on the horizon. Flat vector style, clean shapes, limited palette. Output: exactly one self-contained SVG (viewBox "0 0 800 600", no external references) inside a single ```svg fenced block. Short prose outside the fence is allowed.

Compare with

Evidence and notes

  • No vendor IFEval or GSM8K figures found (Nov 2026); vendor publishes SWE-bench and ARC-AGI-2 instead
  • 14 IFEval prompts unrecoverable under the token budget, counted as failures