The standard Glif tests
We run every model through the same prompts. Open one of these 62 tests to see how each model handled it, side by side.
Image 36
Gibberish prompt
24 models
Inspect how a model interprets gibberish prompts.
Text-to-image · Reliability · Style
Long and detailed prompt
24 models
Adheres to a long and detailed prompt.
Text-to-image · Prompt adherence · Spatial reasoning
Magic laptop
24 models
Creatively merges the ancient/mystical elements with a laptop. Runes are present and the screen shows a correctly themed UI.
Text-to-image · Style · World knowledge
Miniature scene
24 models
Can do tilt-shift scenes. Keeps proportions correct. Correctly produces physically-sound trebuchet. Creates plausible medieval setting.
Text-to-image · Style · World knowledge
Portrait realistic
24 models
Human is realistic and has no artifacts. Applies Kodachrome film effect.
Text-to-image · Style · Reliability
Social media selfie
24 models
Has correct camera positioning. Looks realistic.
Text-to-image · Composition · Reliability
Stacked animals
24 models
Stacks the animals in the correct order on top of each other. Keeps the 1980s photo look and the German medieval town setting.
Text-to-image · Spatial reasoning · Prompt adherence · Style
Tall forehead
24 models
Renders the exaggerated, very tall forehead. No double eyes. Adheres to 1920s black-and-white rubber hose cartoon style.
Text-to-image · Prompt adherence · Style
Trading card
24 models
Produces a plausible trading-card layout with legible text. Anime style.
Text-to-image · Composition · Typography · Style
Edit parts of image
21 models
Cybernetics should be perfectly integrated into the subject. Likeness should be preserved. Legible text and annotations.
Image-to-image · Image editing · Consistency · Typography
Product placement
21 models
Preserves the reference can exactly (label, colors, proportions). Composes a believable bench scene in Ghent. The can should be sized proportionally.
Image-to-image · Reference following · World knowledge
Cat in space
20 models
Adheres to all elements in the prompt. Physically-sound space scene.
Text-to-image · Prompt adherence · World knowledge
Edit from markup (garden)
20 models
Edits only the marked up area. Removes markup. Knows plant names and appearances.
Surgical image editing · Image editing · World knowledge
CMYK texture
20 models
Designs an interesting trading card and applies realistic CMYK dot pattern texture. Applies the texture with the right scaling.
Image-to-image · Reference following · Composition
Character sheet reference
19 models
Places the exact same character and using the same style in a scene.
Character consistency · Reference following · Consistency · Spatial reasoning
Magazine cover
19 models
Uses the exact same character and follows the brief. Good typography and layout.
Character consistency · Consistency · Typography · Composition
Compose image
18 models
Combines the referenced woman, white coat and handbag into one coherent photorealistic fashion shot. Identity and products match the references.
Character consistency · Reference following · Consistency
Style references
17 models
Draws a cat cyborg in the style of the reference images, combining features from all the reference images.
Multi-image reference · Reference following · Style
Beer closeup condensation
16 models
Preserves the exact can design from the reference (label, colors, red 'glif' wordmark) at macro distance. Condensation and droplets look physically real; wordmark stays crisp and legible.
Product photography · Reference following · Consistency
Beer ad text
15 models
On-image text carries the signal: crisp headline and 'glif' wordmark, clean cream-and-red ad layout with negative space. The can itself is preserved exactly as the hero object.
Brand and text rendering · Typography · Reference following · Composition
Beer festival
14 models
Every visible can matches the reference exactly with legible labels. Believable golden-hour festival scene, crowd and stage softly blurred, natural hands and faces.
Product photography · Reference following · Composition · Consistency
Text rendering 2
7 models
Has black, blue and yellow tints. 3D object. Correct lettering and good typography. All info is legible and present.
Text-to-image · Typography · Style
Upscale concrete floor
5 models
Preserves and sharpens the real-world concrete texture without (VAE) artifacts or invented detail.
Image upscaling · Image editing · Reliability
Upscale lowpoly
5 models
Keeps the low-poly game look; does not hallucinate realistic detail.
Image upscaling · Image editing · Style
Upscale painting
5 models
Preserves the watercolor texture; does not turn the painting photographic.
Image upscaling · Image editing · Style
Upscale portrait
5 models
Adds convincing detail without changing the person's identity. Face and anatomy stay correct.
Image upscaling · Image editing · Consistency
Remove background
3 models
Clean matte around the hair/curls. Fully transparent background, no leftover fringes.
Remove background · Image editing · Reliability
Outpaint asymmetric
1 models
Extends 1024px to the right, revealing more of the room consistently. No visible seam.
Image outpainting · Image editing · Composition
Outpaint uniform zoomout
1 models
Zooms out evenly on all four sides. Scene stays coherent; no seams or repetition.
Image outpainting · Image editing · Composition
Outpaint vertical extend
1 models
Extends the photo so that more of the man is visible. Extended part should be consistent with the original photo.
Image outpainting · Image editing · Composition
Outpaint widen portrait
1 models
Extends left and right into a landscape with no visible seams. New content matches lighting and style.
Image outpainting · Image editing · Composition
Segment beer can
1 models
One mask covers the whole can (body, top rim, bottom) with tight edges and none of the white background. The overlay is drawn cleanly over the original image.
Image segmentation · Prompt adherence · Image editing
Segment person curly hair
1 models
The mask covers the whole person including individual curls, with no halo into the background and no eaten hair. Nothing else in the image is masked.
Image segmentation · Prompt adherence · Image editing
Text removal beer can
1 models
Every piece of text is gone: the red 'glif' wordmark, 'Average GLIF PILS', the ring of text at the top, '330 ml' and 'ALC. 5,2% VOL.'. The hop mascot, can shape, colors and lighting are untouched, with clean fills where the text was.
Text removal · Image editing · Reliability
Video 18
iPhone handheld shot
31 models
Handheld phone-selfie camera motion. Looks realistic; the character stays consistent.
Image-to-video · Spatial reasoning · Consistency · Reliability
Miniature diorama
31 models
Keeps the tilt-shift miniature look while adding coherent motion. Motions are physically accurate.
Image-to-video · Style · World knowledge
Animate portrait
27 models
Animates the person while preserving identity. Ends on a questioning expression; motion is coherent.
Image-to-video · Consistency · Prompt adherence · Reliability
Animate radar
27 models
Animates the radar module in a realistic way.
Image-to-video · World knowledge · Reliability
Face likeness
27 models
The person shown at the end is recognizably the same as in the input image. Temporally stable through the bus pass.
Image-to-video · Consistency · Reliability
Action scene
23 models
Smooth camera follows the ollie trick. Motion and physics look physcially accurate.
Text-to-video · Spatial reasoning · World knowledge · Reliability
Comic book history style
23 models
Produces a 4-panel comic layout telling a history-of-the-world narrative. Text and layout are legible. Style and story are consistent throughout the panels.
Text-to-video · Composition · Typography · Consistency
Frame to frame transition watch
20 models
Interpolates smoothly from the first to the last frame. The legs are spinning correctly. The watch is spinning in the correct direction.
First-to-last frame · Reference following · Spatial reasoning
Ref video character kitchen
8 models
The character keeps its exact design and art style from the reference sheet throughout. The kitchen is isometric, the actions happen in order and the motion is smooth.
Multi-image reference · Reference following · Consistency · Style
Video to video (anime style)
8 models
Turns the man into an anime character while keeping his identity and the original motion. Temporally consistent.
Video-to-video · Image editing · Style · Consistency
Ref video fashion walk
5 models
Takes the right element from each reference: her likeness, the white coat and the handbag all match the images and stay consistent for the whole clip. Natural runway walk, no morphing.
Multi-image reference · Reference following · Consistency · Reliability
Audio driven shepherd
1 models
The clip is driven by the provided audio: lips match the words, delivery feels natural, and the shepherd keeps his identity and river setting.
Audio-to-video · Reference following · Consistency · Reliability
Extend ghost alley
1 models
The continuation starts seamlessly from the final frame. The ghost, alley, graffiti and lighting stay consistent, the drift away reads clearly, and any generated audio fits the scene.
Extend video · Consistency · Reliability · Prompt adherence
Keyframes car orbit return
1 models
The trajectory looks like plausible camera movement. No glitches.
Keyframes-to-video · Reference following · Spatial reasoning · Reliability
Keyframes watch pinned endpoints
1 models
Two keyframes pinned to frame 0 and frame 120 (start and end of a 5s clip at 24fps). Interpolates smoothly, legs spinning, correct rotation direction. Comparable to the first-to-last-frame watch test on other models.
Keyframes-to-video · Reference following · Spatial reasoning
Motion control character
1 models
Transfers the reference video's motion onto the character while preserving the character's identity.
Motion control · Reference following · Consistency
Subtitles glass
1 models
Dynamic preset (2x cost): words animate in sync with speech, karaoke-style highlighting. Glass / blurred-backdrop style. Subtitles match the audio.
Video subtitling · Typography · Reliability
Subtitles simple
1 models
Basic preset (1x cost): one static caption block per phrase. Minimal, clean style. Subtitles match the audio.
Video subtitling · Typography · Reliability
Audio 8
Narration calm
7 models
Calm, natural narration of the text in the sample voice. No artifacts.
Text-to-speech · Style · Reliability
Ambient peaceful
3 models
Calm, meditative ambient music with soft piano and strings. Lyrics (if sung) match the lyricsPrompt.
Text-to-music · Style · Prompt adherence
Instrumental cinematic
3 models
Epic orchestral score building to a climax. Strictly instrumental - no vocals.
Text-to-music · Style · Prompt adherence
Structured lyrics rock
3 models
Follows the [Intro]/[Verse]/[Chorus]/[Bridge]/[Outro] structure with the given lyrics. Anthemic stadium-rock sound.
Text-to-music · Prompt adherence · Style
Separate vocals speech
2 models
Returns only the isolated voice: the speech is intact and intelligible with no background noise or room tone left, and no artifacts on the words.
Audio separation · Prompt adherence · Reliability
Nature rain
1 models
Sounds like gentle rain falling on leaves. Realistic; no music or artifacts.
Text-to-sound effects · World knowledge · Reliability
Sfx ghost alley
1 models
Generated sound matches the visuals: ambience fits a night alley, effects line up with the ghost's movements, no music and no speech.
Video-to-audio · World knowledge · Reliability
Speech to speech (ElevenLabs)
1 models
Speaks the input in the target voice. Natural delivery; content unchanged.
Speech-to-speech · Reference following · Reliability