Pixel Football Game Art: a 24-image Qwen, Z-Image, FLUX.2, and Anima benchmark
This benchmark asks a production question: which local image model can provide useful source references for a portrait football game? It does not rank generic beauty. Each of four models generated two independent images for three asset roles—character, environment, and one gameplay prop—giving 24 full results. Every image below links to its full-size optimized file. 中文版
The practical result is clear. Qwen Image 2512 FP8 is the strongest overall source generator and the only model that repeatedly delivered convincing authored pixel clusters. FLUX.2 Klein is the strongest alternative for environment layout and prop geometry, but its output often reads as vector or 2.5D art. Z-Image Turbo is fast but unreliable for this exact art direction. Anima understands illustration and anatomy, yet usually produces manga art rather than pixel art.
These are source references, not runtime sprites, production backgrounds, turnarounds, or ready-to-import game assets.
Test design
All jobs ran serially in managed ComfyUI on an RTX 5090 Laptop GPU with 24,463 MiB VRAM. SageAttention was disabled. Seeds were 42 and 1337; there were no LoRAs, ControlNets, contact sheets, or hidden retries. The role buckets were character 1056 × 1584, background 928 × 1664, and prop 1328 × 1328.
| Installed variant | Native workflow used | Six-run mean | Peak VRAM |
|---|---|---|---|
| Qwen Image 2512 official FP8 e4m3fn | AuraFlow shift 3.1, Euler/simple, 50 steps, CFG 4.0 | 212.044 s | 23,981 MiB |
| Z-Image Turbo official BF16 | AuraFlow shift 3.0, Euler/simple, 9 steps, CFG 1.0, empty negative | 22.422 s | 24,015 MiB |
| FLUX.2 Klein 9B community Q8_0 GGUF | UniPC/simple, 12 steps, FluxGuidance 3.5, sampler CFG 1.0 | 51.811 s | 22,062 MiB |
| Anima Aesthetic v1.1 official BF16 | AuraFlow shift 3.0, ER-SDE/simple, 30 steps, CFG 4.0 | 36.289 s | 8,103 MiB |
The FLUX checkpoint must be read literally: flux-2-klein-9b-Q8_0.gguf, SHA-256 dfa8908d…fe5297. It is the installed community Q8_0 quantization, not the originally specified official FP8 file. Qwen used qwen_image_2512_fp8_e4m3fn.safetensors; Z used z_image_turbo_bf16.safetensors; Anima used anima-aesthetic-v1.1.safetensors. Full file sizes, hashes, prompts, graph JSON, timings, and validation receipts are retained in the evaluation archive.
Official references: Qwen Image 2512 ComfyUI workflow, Z-Image repository, FLUX.2 Klein model card, FLUX.2 ComfyUI workflows, and Anima model card. Anima was also checked against the official ComfyUI template and a reproducible community workflow.
Role 1: isolated character reference
The brief asked for one original adult football protagonist, fully visible, in a three-quarter-left running kick with exactly one ice-white ball. The target visual language was a restricted dark-teal/cyan/coral palette with crisp square clusters and stepped diagonals.
Qwen Image 2512 FP8
Both are accepted source references. The silhouette, anatomy, ball, and costume motif survive thumbnail viewing; their cluster shapes look deliberately placed rather than applied as a coarse filter.
Z-Image Turbo BF16
The figure remains readable, but seed 42 turns the football into a malformed oval and seed 1337 omits it. Large blocks imitate low resolution without establishing strong pixel clusters. Both are rejected as final pixel references.
FLUX.2 Klein 9B Q8_0 GGUF
The anatomy and ball are clean in both seeds, making them useful composition references. The finish is smoother and more vector-like than native pixel art, so neither should be mistaken for a production sprite.
Anima Aesthetic v1.1
Seed 42 has a floating footwear/ball ambiguity; seed 1337 has clean anatomy but is halftone manga illustration. The latter is usable for a broad character idea, not as evidence that Anima produced pixel art.
Role 2: football environment reference
The environment brief specified an empty portrait player tunnel opening onto a night pitch and one distant goal, with dark top and bottom zones reserved for real UI. It excluded players, balls, text, scoreboards, and interface elements.
Qwen Image 2512 FP8
These have the best combination of tunnel depth, distant target, palette, and authored edge treatment. Seed 42 includes a rectangular wall panel that can be mistaken for a scoreboard, while smooth lighting keeps both from being perfectly strict pixel art.
Z-Image Turbo BF16
Both images are legible stadium scenes, but neither preserves the requested player-tunnel camera. They are partial stadium mood references, not passes for this layout task.
FLUX.2 Klein 9B Q8_0 GGUF
FLUX produces the strongest tunnel construction and UI-safe zoning. Seed 1337 contains tiny colored marks that resemble unintended signage, and both frames lean toward flat vector or low-poly-like shapes rather than authored pixel clusters.
Anima Aesthetic v1.1
Anima understands tunnel depth and dark composition, especially in seed 42, but renders manga line work and smooth tonal areas. Seed 1337 also loses the distant goal. These are partial layout sketches, not pixel backgrounds.
Role 3: one gameplay prop reference
The prop was one split gold goal-shield obstacle: two asymmetric shield halves, one small open goal, and four cyan shard clusters. It needed a bold silhouette readable after nearest-neighbour reduction to 64 × 64.
Qwen Image 2512 FP8
Both passes isolate one readable object and clearly separate shield, goal, split, and shards. This is the strongest pair for later 2D redraw and the safest pair for selecting a 3D-reconstruction input after approval.
Z-Image Turbo BF16
These are low-poly 3D renders, not pixel art. Seed 1337 still offers a coherent geometry sketch, but the pair fails the stated visual language and should be labelled geometry reference only.
FLUX.2 Klein 9B Q8_0 GGUF
Both have excellent part separation, silhouette, and construction cues. They are useful 2.5D/3D geometry references, but the vector-like finish does not satisfy native pixel-art credibility.
Anima Aesthetic v1.1
Seed 42 is a coherent but flat side-biased shield-and-goal concept. Seed 1337 becomes a multi-object diorama with field, goal, cube, and many shards, violating the single-prop requirement; it is rejected.
Scores by production role
Each cell is the mean of two seeds and five 1–5 criteria: role readability, brief compliance, pixel credibility, artifact control, and downstream usefulness. The overall value is the mean of all 30 criterion scores per model—not a universal image-quality score.
| Model | Character | Environment | Prop | Overall |
|---|---|---|---|---|
| Qwen Image 2512 FP8 | 5.00 | 4.60 | 5.00 | 4.87 |
| FLUX.2 Klein 9B Q8_0 | 4.40 | 4.30 | 4.40 | 4.37 |
| Z-Image Turbo BF16 | 2.60 | 3.60 | 3.30 | 3.17 |
| Anima Aesthetic v1.1 | 2.80 | 3.70 | 2.70 | 3.07 |
Exact prompts
The semantic target stayed common, but wording and negative-prompt policy followed each model's native convention. Seeds and dimensions are listed above; the complete graph JSON for every sample is archived.
Character prompts — all four models
Qwen: Create one original adult football protagonist for Kickshatter: Pixel Cup, full body and fully inside a portrait action plate, in a clear three-quarter-left running-kick pose with exactly one ice-white football in front of the planted boot. The player has warm brown skin, a compact athletic build, short dark-auburn high-top hair with one small cyan forelock, a sponsor-free midnight-teal shirt with a single coral diagonal shard motif, dark teal shorts and socks, and pale ice boots. This must read as native 2D pixel-game art at the same intentional authored scale as a 232 by 312 character plate: crisp hard square pixel clusters, stepped diagonal edges, hand-placed-looking highlight ramps, selective deep ink outlines, and a restricted palette of Ink #020A12, Tunnel #071723, cyan, mint, gold, coral, ice-white, and warm-brown skin. Use a flat deep-ink field with only a few short cyan and coral speed strokes; leave no frame, card, typography, logo, UI, crowd, stadium architecture, or extra object. The head, kicking leg, planted leg, both boots, and the one ball must remain legible at a 25 percent thumbnail crop.
Qwen negative: photorealistic, 3D render, low-poly, isometric, CGI, smooth vector, glossy material, smooth gradient, airbrush, antialiasing, blur, poster, trading card, border, text, letters, numbers, sponsor, logo, watermark, real player likeness, club, flag, duplicate ball, second person, cropped limbs, malformed anatomy
Z-Image Turbo: One original adult football protagonist for Kickshatter: Pixel Cup, full body in a clear three-quarter-left running-kick pose with exactly one ice-white football ahead of the planted boot. Warm brown skin, compact athletic build, short dark-auburn high-top hair with one small cyan forelock, sponsor-free midnight-teal shirt with one coral diagonal shard, dark teal shorts and socks, pale ice boots. Native 2D pixel-game action plate at the authored scale of a small 232 by 312 character sprite: deliberate hard square pixel clusters, stepped diagonal edges, selective deep-ink outlines and highlight ramps, restricted ink, tunnel blue, cyan, mint, gold, coral, ice-white and warm-brown palette. Flat deep-ink field with only a few cyan and coral speed strokes, no frame, text, UI, logo, crowd, scenery, second subject or second ball; head, kicking leg, planted leg and both boots readable at 25 percent thumbnail size. Negative: empty.
FLUX.2 Klein: One original adult football protagonist for Kickshatter: Pixel Cup, full body in a clear three-quarter-left running-kick pose with one ice-white football ahead of the planted boot. Subject: warm-brown skin, compact athletic build, short dark-auburn high-top hair with one small cyan forelock, sponsor-free midnight-teal shirt with one coral diagonal shard, dark teal shorts and socks, pale ice boots. Action: a readable run-up and kick, both boots and the ball clear at a 25 percent thumbnail. Style: native 2D pixel-game action plate at the authored scale of a 232 by 312 character sprite, deliberate hard square pixel clusters, stepped diagonal edges, deep-ink outlines, hand-placed highlight ramps, restricted ink, tunnel blue, cyan, mint, gold, coral, ice-white and warm-brown palette. Context: a flat deep-ink field with a few cyan and coral speed strokes, no frame, card, lettering, logo, UI, crowd, architecture or extra subject. Negative: empty.
Anima: masterpiece, best quality, safe, 1boy, original football protagonist, full body, three-quarter left running kick, one ice white football ahead of the planted boot, warm brown skin, compact athletic build, short dark auburn high top hair, cyan forelock, sponsor-free midnight teal shirt, coral diagonal shard motif, dark teal shorts, dark teal socks, pale ice boots, native 2D pixel game art, 232 by 312 action plate scale, hard square pixel clusters, stepped diagonal edges, selective deep ink outlines, hand placed highlight ramps, restricted ink tunnel blue cyan mint gold coral ice white warm brown palette, flat deep ink field, cyan and coral speed strokes, head kicking leg planted leg both boots and ball readable at 25 percent thumbnail
Anima negative: worst quality, low quality, score_1, score_2, score_3, artist name, blurry, jpeg artifacts, photorealistic, 3D render, low-poly, isometric, smooth vector, glossy material, smooth gradient, airbrush, antialiasing, poster, trading card, border, text, letters, numbers, sponsor, logo, watermark, real player likeness, club, flag, duplicate ball, second person, cropped limbs, malformed anatomy
Environment prompts — all four models
Qwen: Create one empty portrait menu-background plate for Kickshatter: Pixel Cup. Look straight out from a deep Ink #020A12 player tunnel through one angular cut-corner opening to a quiet night football pitch and one small ice-white distant goal. The tunnel has dark Tunnel #071723 walls, deliberately stepped edges, a few tiny gold practical lights, cyan floodlight rims, deep blue stadium tiers, sparse square crowd-light clusters, and only one narrow pitch-green stripe at the far end. This must be a native 2D pixel-game background at the same authored scale as a 420 by 912 logical-point portrait screen: deliberate hard pixel clusters, stair-step perspective, discrete dithering only where needed, no blur, no smooth gradients, no glossy 3D materials. Keep the top and bottom dark enough for native game UI, while the central opening and goal stay readable. No players, ball, obstacle, menu, labels, text, logos, scoreboard, flags, camera, equipment, watermark, border or UI.
Qwen negative: photorealistic, 3D render, low-poly, isometric, CGI, smooth vector, smooth gradient, soft blur, bloom, lens flare, poster, screenshot UI, typography, letters, numbers, logo, watermark, player, person, football, obstacle, crowd faces, flag, camera, tripod, scoreboard
Z-Image Turbo: One empty portrait menu-background plate for Kickshatter: Pixel Cup. Centered view from inside a deep-ink player tunnel through one angular cut-corner opening to a quiet night football pitch and one small ice-white distant goal. Dark tunnel-blue walls, stepped pixel edges, a few tiny gold practical lights, cyan floodlight rims, deep-blue stadium tiers, sparse square crowd-light clusters and one narrow pitch-green stripe at the far end. Native 2D pixel-game background at a 420 by 912 screen scale: hard pixel clusters, stair-step perspective, restrained discrete dithering, a dark quiet top and bottom for real UI, a readable opening and goal. No player, football, obstacle, menu, text, logo, scoreboard, flag, camera, equipment, border or UI. Negative: empty.
FLUX.2 Klein: One empty portrait menu-background plate for Kickshatter: Pixel Cup. Subject: a centered view from within a deep Ink #020A12 player tunnel, through one angular cut-corner opening, toward a quiet night football pitch and one small ice-white distant goal. Style: native 2D pixel-game background at the authored scale of a 420 by 912 logical-point screen, hard pixel clusters, stepped perspective, minimal discrete dithering, dark tunnel-blue walls, cyan floodlight rims, a few gold practical lights, deep-blue stadium tiers, sparse square crowd-light clusters, one narrow pitch-green stripe. Context: upper and lower frame remain dark and quiet for the real UI; no player, ball, obstacle, menu, lettering, logo, scoreboard, flag, camera, equipment, border or interface. Negative: empty.
Anima: masterpiece, best quality, safe, original empty portrait menu background for a pixel football game, centered view from inside a deep ink player tunnel through one angular cut-corner opening to a quiet night football pitch and one small ice white distant goal, dark tunnel blue walls, hard pixel clusters, stair step perspective, a few tiny gold practical lights, cyan floodlight rims, deep blue stadium tiers, sparse square crowd light clusters, one narrow pitch green stripe, dark quiet top and bottom UI safe areas, native 2D pixel game background at 420 by 912 screen scale
Anima negative: worst quality, low quality, score_1, score_2, score_3, artist name, blurry, jpeg artifacts, photorealistic, 3D render, low-poly, isometric, CGI, smooth vector, smooth gradient, soft blur, bloom, lens flare, poster, screenshot UI, typography, letters, numbers, logo, watermark, player, person, football, obstacle, crowd faces, flag, camera, tripod, scoreboard, border, interface
Prop prompts — all four models
Qwen: Create one original split gold goal-shield obstacle for the 2D pixel-football puzzle Kickshatter: Pixel Cup. Use one readable front three-quarter gameplay view. It is a thick angular gold shield wall split into two asymmetric halves by one diagonal impact line; the opening reveals one small ice-white football goal with a dark-navy square net. Four large cyan glass shard clusters are falling from the split. This is a small 2D gameplay prop, not a model sheet or product render: give it one bold hard silhouette, deliberate square pixel clusters, stair-step corners, deep Ink #020A12 outline, three flat gold highlight ramps, ice-white goal frame, cyan shards, dark net pixels and a very low pitch-green strip under a flat deep-ink background. The split, both gold halves, the open goal and cyan shards must remain immediately readable after a nearest-neighbour reduction to 64 by 64 pixels. No player, football, stadium, pedestal, studio ground, title, number, logo, text, border, HUD, soft shadow, blur, smooth gradient, isometric view, low-poly render, photoreal material, CGI or vector finish.
Qwen negative: photorealistic, 3D render, low-poly, isometric, CGI, smooth vector, smooth gradient, glossy material, product render, studio scene, airbrush, blur, antialiasing, text, letters, numbers, logo, watermark, player, football, stadium, crowd, card frame, UI, contact sheet, duplicate prop
Z-Image Turbo: One original split gold goal-shield obstacle for Kickshatter: Pixel Cup in a readable front three-quarter gameplay view. A thick angular gold shield wall splits into two asymmetric halves along one diagonal impact line and reveals one small ice-white open football goal with a dark-navy square net; four cyan glass shard clusters fall from the split. Small native 2D pixel-game prop: bold hard silhouette, deliberate square pixel clusters, stair-step corners, deep-ink outline, three flat gold highlight ramps, ice-white goal frame, cyan shards, dark net pixels, very low pitch-green strip under a flat deep-ink background. The split, gold halves, open goal and cyan shards must read at 64 by 64 pixels. No player, ball, stadium, pedestal, studio ground, text, logo, UI, smooth shading, product render or extra prop. Negative: empty.
FLUX.2 Klein: One original split gold goal-shield obstacle for Kickshatter: Pixel Cup, viewed from a readable front three-quarter gameplay angle. Subject: one thick angular gold shield wall, split into two asymmetric halves around one small ice-white open football goal with a dark-navy square net; four cyan glass shard clusters fall from one diagonal impact line. Action: both shield halves pull apart and reveal the goal. Style: a small native 2D pixel-game prop, bold hard silhouette, deliberate square pixel clusters, stepped corners, deep-ink outline, three flat gold highlight ramps, ice-white goal frame and cyan shard highlights. Context: flat deep-ink background, one very low pitch-green strip, no studio ground, no 3D material, no player, ball, stadium, text, logo, UI or extra prop; the split, goal and shards must read at 64 by 64 pixels. Negative: empty.
Anima: masterpiece, best quality, safe, original split gold goal shield obstacle for a 2D pixel football puzzle, one readable front three-quarter gameplay view, thick angular gold shield wall split into two asymmetric halves by one diagonal impact line, one small ice white open football goal with a dark navy square net, four large cyan glass shards falling from the split, small native 2D pixel game prop, bold hard silhouette, deliberate square pixel clusters, stair-step corners, deep ink outline, three flat gold highlight ramps, ice-white goal frame, dark net pixels, low pitch green strip, flat deep ink background, readable at 64 by 64 pixels
Anima negative: worst quality, low quality, score_1, score_2, score_3, artist name, blurry, jpeg artifacts, photorealistic, 3D render, low-poly, isometric, CGI, smooth vector, smooth gradient, glossy material, product render, studio scene, airbrush, antialiasing, text, letters, numbers, logo, watermark, player, football, stadium, crowd, card frame, UI, contact sheet, duplicate prop
Validation and limitations
All 24 source PNGs passed full decoding, expected-dimension checks, manifest SHA-256 comparison, duplicate detection, and full-size visual review. Their SHA-256 values are all unique. The published set contains native-size, 800-pixel, and 480-pixel WebP and AVIF derivatives with intrinsic dimensions and lazy loading; every derivative was decoded and dimension-checked, and native-size derivatives were checked against the source image in pixel space.
Two seeds expose obvious repeatability problems, but they do not establish a universal leaderboard. Model-native wording also means the prompts are semantically aligned rather than character-for-character identical. Anima and FLUX.2 Klein 9B weights carry non-commercial licenses; these outputs remain evaluation references, and production use requires a separate license review. Finally, an attractive render is not a benchmark pass: Z's and FLUX's 3D-looking props are labelled honestly rather than called pixel art.
For this game, I would take Qwen's seed-42 character, Qwen's seed-1337 environment, and Qwen's seed-42 prop into a controlled editing evaluation. FLUX's prop pair is the better alternate when geometric construction matters more than strict pixel language. None of these files goes directly into the game.