v0.3: 6 steps by default, new 9-step setting (7 turbo + 2 base-model steps) (part 2)

.gitattributes

@@ -222,3 +222,15 @@ compare/img/case18/turbo9.webp filter=lfs diff=lfs merge=lfs -text
 compare/img/case19/turbo9.webp filter=lfs diff=lfs merge=lfs -text
 compare/img/case20/turbo9.webp filter=lfs diff=lfs merge=lfs -text
 compare/img/case21/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/case22/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/case23/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/case24/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/case25/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/en_1/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/en_2/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/en_3/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/en_3/turbo9_t.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/en_4/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/en_4/turbo9_t.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/zh_1/turbo9.webp filter=lfs diff=lfs merge=lfs -text
+compare/img/zh_7/turbo9.webp filter=lfs diff=lfs merge=lfs -text

compare.py

@@ -1,5 +1,5 @@
 """The demo's Comparison tab: 32 of the 37 examples of the official Qwen/Qwen-Image-2.1 Space, rendered with viggle-turbo
-v0.2.1 in 6 steps (and in 8 steps on the 5 dense-text examples) and with the base model in 40 steps. Both get the same
+v0.3 in 6 steps and in 9 steps (7 turbo steps + 2 base-model steps) and with the base model in 40 steps. Both get the same
 prompt, input images and seed 42, with prompt enhancement off. No GPU: everything is pre-rendered. compare/ (WebP images
 + cases.json) is written by release/build_compare_space.py, and upload_space.sh builds it into the Space."""
 import json
@@ -18,20 +18,19 @@ KINDS = {
     "Editing": lambda case: len(case["inputs"]) > 0,
     "Multi-image editing": lambda case: len(case["inputs"]) > 1,
     "Text-to-image": lambda case: not case["inputs"],
-    "Dense text (+8 steps)": lambda case: "turbo8" in case,
+    "Dense text": lambda case: case["dense"],
 }
 
 
 def best_turbo(case):
-    """The turbo render an example opens on: 8 steps where it has one (the dense-text examples), else 6."""
-    return "turbo8" if "turbo8" in case else "turbo"
+    """The turbo render an example opens on: 9 steps on the dense-text examples, else 6."""
+    return "turbo9" if case["dense"] else "turbo"
 
 
 def arms(case):
     """(menu label, key) for every render of an example; the render time is in the label."""
     menu = [(f"viggle-turbo · 6 steps · {case['turbo_s']:.1f} s", "turbo")]
-    if "turbo8" in case:
-        menu.append((f"viggle-turbo · 8 steps · {case['turbo8_s']:.1f} s", "turbo8"))
+    menu.append((f"viggle-turbo · 9 steps (7 + 2 base) · {case['turbo9_s']:.1f} s", "turbo9"))
     menu.append((f"Qwen-Image-2.1 base · 40 steps · {case['base_s']:.1f} s", "base"))
     if case["reference"]:
         menu.append(("Qwen API reference (bundled with Qwen's Space)", "reference"))
@@ -70,9 +69,9 @@ def render():
         "[Qwen/Qwen-Image-2.1 Space](https://huggingface.co/spaces/Qwen/Qwen-Image-2.1). Both models get the same prompt, "
         "input images and seed, with prompt enhancement off, one sample each and no seed picking. "
         f"At **{speedup:.1f}× less time** the turbo is very competitive with the base model; the clearest gaps are "
-        "small, dense text (8 steps narrows it) and complicated edits, where the base is still ahead. "
-        "Pick an example on the left, then drag the slider. The 5 dense-text examples open on the 8-step turbo (6 steps is in "
-        "the menu), and the menus also offer Qwen's own API reference where the Space ships one.\n\n"
+        "small, dense text (9 steps often narrows it) and complicated edits, where the base is still ahead. "
+        "Pick an example on the left, then drag the slider. The 5 dense-text examples open on the 9-step turbo, the others on "
+        "6 steps (both are in every menu), and the menus also offer Qwen's own API reference where the Space ships one.\n\n"
         f"Median time, editing: **{median(edits, 'turbo_s'):.1f} s vs {median(edits, 'base_s'):.1f} s** · "
         f"text-to-image (~4 MP): **{median(t2is, 'turbo_s'):.1f} s vs {median(t2is, 'base_s'):.1f} s** · "
         f"all {len(CASES)} examples: **{speedup:.1f}× faster** (one NVIDIA B200, one pipeline call each)"
@@ -97,13 +96,11 @@ def render():
                                           interactive=False, show_copy_button=True, scale=2)
     with gr.Accordion("How these were made", open=False):
         gr.Markdown(
-            "- **Turbo:** Qwen-Image-2.1 + the viggle-turbo v0.2.1 LoRA (rank 256), 6 steps on sigmas "
+            "- **Turbo:** Qwen-Image-2.1 + the viggle-turbo v0.3 LoRA (rank 256), 6 steps on sigmas "
             "`[1, 0.9375, 0.875, 0.75, 0.5, 0.25]`, no classifier-free guidance, exactly as the Generate tab runs it.\n"
-            "- **Turbo, 8 steps** (the 5 dense-text examples): sigmas `[1, 0.9375, 0.875, 0.75, 0.625, 0.5, 0.25, 0.125]`. "
-            "At 6 steps small Latin text can print twice, like a double exposure (decided in the 0.75 → 0.5 step), and small "
-            "Chinese strokes can get colour blotches (the last 0.25 → 0 step); each added sigma splits one of the two. In our "
-            "OCR tests 8 steps raise word recall from 0.71 to 0.83 on the academic infographic's caption (16 seeds) and from "
-            "0.76 to 0.90 on the architecture board's Chinese labels (4 seeds); the base model scores 0.95 on both.\n"
+            "- **Turbo, 9 steps:** the Generate tab's 9-step setting: 7 turbo steps, then the LoRA is switched off and the base "
+            "model takes the last two. Finer detail than 6 steps and small text right more often; about 1.4–1.5× the time of "
+            "6 steps.\n"
             "- **Base:** Qwen-Image-2.1, 40 steps, default scheduler, `true_cfg_scale=1.0` (the pipeline default).\n"
             "- **Both:** diffusers `QwenImage21Pipeline` in bf16, seed 42. Input images are encoded at 1024² pixels; the output keeps "
             "the aspect ratio of Qwen's reference at 2048² pixels for text-to-image and 1536² for editing, in multiples of 32.\n"
@@ -118,7 +115,7 @@ def render():
         )
 
     def show(case_id, left_key, right_key, fixed="left"):
-        """Keeps the chosen pair when the new example has it (turbo8 and reference are per-example), else turbo vs base.
+        """Keeps the chosen pair when the new example has it (reference is per-example), else turbo vs base.
         If both menus land on the same render, the one the user just changed (`fixed`) wins and the other moves off it."""
         case = BY_ID[case_id]
         keys = [key for _, key in arms(case)]
@@ -134,7 +131,7 @@ def render():
     def pick(ids, left_key, right_key, evt: gr.SelectData):
         """A new example keeps the chosen pair, except that a turbo menu moves to that example's best turbo render."""
         case = BY_ID[ids[evt.index]]
-        turbo = lambda key: best_turbo(case) if key in ("turbo", "turbo8") else key  # noqa: E731
+        turbo = lambda key: best_turbo(case) if key in ("turbo", "turbo9") else key  # noqa: E731
         return show(case["id"], turbo(left_key), turbo(right_key))
 
     def filter_cases(name):

examples/manifest.json

@@ -7,7 +7,7 @@
    null
   ],
   "result": "qwen_storyboard.png",
-  "info": "Pre-rendered example · seed `42` · 2048×1152 · 6 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `42` · 2048×1152 · 6 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "生成一张无缝的分镜插画，**现代真实偶像剧风格**（电影感写实摄影，柔和自然光与暖色调，浅景深虚化，韩系都市偶像剧质感，画面精致高级）。\n\n【版面（必须严格遵守）】\n- 恰好 6 个分镜格，**等宽**，排成**单独一横排（1 行 6 列）**；不是网格、不是两行，绝不要把任何一格再拆成上下两块。\n- 每格都是**竖长条矩形**（高明显大于宽，近似 9:16 竖条）。\n- 从左到右依次编号 1、2、3、4、5、6，每格左上角一个圆形数字徽标；**每个数字只出现一次，不重复、不跳号、不缺号。**\n- **每一格里都必须画出角色本人**在做该格的动作，角色占该格画面主体、完整清晰（半身或全身，约占该格高度 2/3 以上）；**严禁出现只有背景、没有人物的空格。**\n- 6 格之间画风、光影氛围、人物一致。输出为一张扁平整图，6 格并排合并，不要输出多张分离文件。\n\n【角色】以输入参考图（含面部特写与正/侧/背三视全身立绘）为角色设计基准：现代年轻女性，棕色波浪长发带空气刘海，妆容清透，穿粉色方领修身长袖针织上衣、棕色灯芯绒过膝半裙配棕色皮带、棕色皮质机车靴。每一格里都是同一个人——面容、发型、服饰、身材比例保持一致，只改变姿势、表情、动作与所处场景；参考图仅定义外观，不要照搬其站姿。把角色自然地融入各自的真实场景中，透视比例正确、光影贴合。\n\n【逐格场景与内容】\n第1格：清晨阳光洒入的现代简约公寓，女孩站在落地窗边，双手捧着马克杯，望向窗外的城市天际线，神情恬静。\n第2格：白天繁华的都市街道人行道，两旁是时尚店铺橱窗和梧桐树，女孩背着单肩包边走边微微回头，浅浅微笑。\n第3格：文艺清新的咖啡馆靠窗卡座，女孩坐着低头看手机，嘴角带笑，窗外光线柔和，桌上放着一杯拿铁。\n第4格：暖色调的独立书店书架走廊，女孩踮脚从书架上抽出一本书，侧脸专注，光线温柔。\n第5格：傍晚金色晚霞下的城市人行天桥，女孩倚着栏杆回眸，背景是虚化的车流与高楼，暖光洒在脸上。\n第6格：夜晚霓虹灯光的街角，女孩仰头微笑看向前方，暖黄路灯与散景光斑环绕，都市夜景氛围。",
   "steps": 6,
   "size_label": "16:9 · 2048×1152 (1536² area)",
@@ -22,7 +22,7 @@
    null
   ],
   "result": "qwen_rgba_bride.png",
-  "info": "Pre-rendered example · seed `42` · 1664×2496 · 6 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `42` · 1664×2496 · 6 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "这是一张带有透明度的RGBA格式图像，呈现一位身着华丽婚纱的动漫风格女性角色。画面中无任何可识别的文字信息。角色为年轻女性，拥有白皙肌肤、红润脸颊与明亮红色眼眸，面带温柔微笑，表情甜美而幸福。她头戴精致银色皇冠与白色蕾丝头纱，金色长发自然垂落肩头，发丝柔顺且富有光泽。身穿多层荷叶边设计的粉白渐变抹胸婚纱，裙摆蓬松飘逸，层次丰富，材质轻盈如花瓣般展开；胸前手持一束由白色玫瑰与绿色叶片组成的捧花，花束饱满，绿叶点缀其间，增添清新感。双腿修长，穿着透明薄纱质感的过膝袜与饰有玫瑰图案的白色高跟鞋，姿态优雅站立，双手轻握捧花置于腹前。整体构图为全身正面视角，色彩柔和梦幻，以粉、白、绿为主色调，光影细腻，具有典型的日系二次元插画美学风格，适用于游戏立绘、视觉小说或数字艺术收藏等场景。 该图像具有alpha通道，背景是透明的。",
   "steps": 6,
   "size_label": "2:3 · 1664×2496 (2048² area)",
@@ -37,7 +37,7 @@
    null
   ],
   "result": "t2i_style_file.png",
-  "info": "Pre-rendered example · seed `42` · 2720×1536 · 6 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `42` · 2720×1536 · 6 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "An elegant fashion-magazine style infographic, landscape 16:9: a character style file for an original fictional heroine named Mira Solenne. Warm ivory paper background with a faint watercolor texture, thin champagne-gold hairlines dividing the panels, generous white space, refined high-contrast serif headings and small clean sans-serif labels. Palette: midnight navy, pearl white, silver grey, dusty lilac and champagne gold.\n\nTop banner: a large serif title \"MOONLIGHT SILK\" on the left with a short gold rule beneath it, and on the right in small letter-spaced caps \"VOL.05 · STYLE FILE\".\n\nCenter column (the largest area): a full-body watercolor and ink fashion illustration of Mira Solenne, a young woman with a sleek silver-lilac bob, in a flowing midnight-navy silk slip dress under a cropped pearl-white jacket, silver strappy heels, standing in a graceful pose, a soft lilac watercolor wash and a pale crescent moon behind her. Below her feet, her name in a small italic serif: \"Mira Solenne\".\n\nLeft column, top: a panel headed \"PROFILE\" with four lines: \"Style: Modern Romantic\", \"Height: 172 cm\", \"Colors: Navy & Pearl\", \"Mood: Calm & Dreamy\".\nLeft column, bottom: a panel headed \"DETAILS\" with four small line-art icons in a 2×2 grid, captioned \"Pearl Earrings\", \"Silk Scarf\", \"Moon Clutch\", \"Silver Heels\".\n\nRight column, top: a panel headed \"PALETTE\" with a row of six round color swatches captioned \"Midnight\", \"Pearl\", \"Silver\", \"Lilac\", \"Gold\", \"Ink\".\nRight column, middle: a panel headed \"STYLE RADAR\" with a pentagon radar chart filled in translucent lilac, five axes labeled \"Elegance 92%\", \"Romance 88%\", \"Comfort 76%\", \"Minimal 70%\", \"Edge 64%\".\nRight column, bottom: a panel headed \"3 LOOKS\" with three small watercolor sketches of her: in a navy gown captioned \"Gala Night\", in a white trench coat captioned \"City Walk\", in a lilac dress and straw hat captioned \"Sunday Brunch\".\n\nBottom strip across the full width: five rounded outline keyword chips with gold borders, \"Satin\", \"Moonlit\", \"Soft Tailoring\", \"Quiet Luxury\", \"Night Garden\".",
   "steps": 6,
   "size_label": "16:9 · 2720×1536 (2048² area)",
@@ -52,7 +52,7 @@
    null
   ],
   "result": "qwen_poster.png",
-  "info": "Pre-rendered example · seed `42` · 1664×2496 · 6 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `42` · 1664×2496 · 6 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "Design a refined travel poster for a fictional night train. Render the headline exactly as \"THE MIDNIGHT EXPRESS\" and the subtitle \"A journey under the stars\". A silver train curves through dark blue mountains beneath a crescent moon. Art Deco geometry, ivory and gold lettering, clear typographic hierarchy, generous margins, print-ready composition.",
   "steps": 6,
   "size_label": "2:3 · 1664×2496 (2048² area)",
@@ -67,7 +67,7 @@
    "bird.webp"
   ],
   "result": "edit_3ref_klein.png",
-  "info": "Pre-rendered example · seed `42` · 928×1152 · 6 steps · v0.2.1 LoRA r256 · prompt enhanced",
+  "info": "Pre-rendered example · seed `42` · 928×1152 · 6 steps · v0.3 LoRA r256 · prompt enhanced",
   "used_prompt": "Compose a scene where the person from <image1>, standing beside the vintage brown car, is gently petting the fluffy cat from <image2> that is perched on a stone windowsill with green shutters, while the bird from <image3>, with a red crown and long beak, stands next to them, all under warm cinematic lighting with a shallow depth of field focusing on the trio, keeping the background and subjects' identities from the input images unchanged.",
   "steps": 6,
   "size_label": "Auto · match the last reference (1024² area)",
@@ -82,7 +82,7 @@
    null
   ],
   "result": "qwen_mask_edit.png",
-  "info": "Pre-rendered example · seed `42` · 1248×1888 · 6 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `42` · 1248×1888 · 6 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "图中画圈标注的地方需要补上一名骑坐姿态的西部牛仔男子。他头戴棕色宽檐牛仔帽，蓄着浓密的络腮胡，脸侧向画面左方；上身穿棕色帆布夹克，里面搭配深蓝色牛仔衬衫，颈间系着浅棕色围巾；下身穿着带流苏的棕色皮质护腿，脚穿皮靴踩进马镫，一只手搭在鞍部附近，整体呈现自然的骑乘状态。",
   "steps": 6,
   "size_label": "2:3 · 1248×1888 (1536² area)",
@@ -97,7 +97,7 @@
    null
   ],
   "result": "t2i_launch.png",
-  "info": "Pre-rendered example · seed `3` · 2496×1664 · 8 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `3` · 2496×1664 · 8 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "Create a premium, minimalist launch graphic for an AI image model, designed for a Twitter/X announcement.\n\nAspect ratio: 16:9.\nStyle: elegant, high-end, modern AI product launch visual. Clean Apple / NVIDIA / premium creative-software aesthetic. Dark, cinematic, sophisticated, not flashy or cluttered.\n\nBACKGROUND\n- Deep black to charcoal gradient background.\n- Very subtle blue and violet ambient glow.\n- Soft cinematic lighting, slight glossy reflections near the bottom.\n- Large areas of negative space.\n- No unnecessary particles, grids, icons, decorative UI, or busy textures.\n\nTOP-LEFT TYPOGRAPHY\n\nMain title:\n“Qwen-Image-2.1 ❤️ Viggle Turbo”\n\n- Large bold geometric sans-serif.\n- Clean Swiss-style typography.\n- Bright soft-white text.\n- Red heart between the two model names.\n- Precise kerning and professional spacing.\n- Keep the title on one line.\n- Position around 4–5% from the left and 8–10% from the top.\n\nDirectly underneath, smaller subtitle:\n“DMD-distilled · Generate & Edit”\n\n- Thin or regular sans-serif.\n- Around 30–35% of the main title font size.\n- Soft light-gray / off-white.\n- Generous letter spacing.\n- Minimal and understated.\n\nCENTER HERO MESSAGE\n\nLarge dominant typography:\n\n“6-step”\n“generation”\n\non two lines.\n\n- “6-step” should be extremely large.\n- “generation” slightly smaller but still bold.\n- Heavy geometric sans-serif.\n- White / very subtle cool-white gradient.\n- Tight line spacing.\n- Perfectly clean typography.\n- Position slightly above the vertical center.\n- The text should be the strongest visual element in the design.\n- No exaggerated 3D text, extrusion, chrome effects, or heavy shadows.\n- Only a very subtle soft glow.\n\nVISUAL ELEMENT\n\nCreate a single elegant curved ribbon of generated images flowing across the lower half of the composition.\n\nThe ribbon should:\n- Start from the lower-left foreground.\n- Curve smoothly toward the center-right.\n- Continue upward toward the upper-right background.\n- Feel like a sophisticated cinematic film strip or flowing image-generation sequence.\n- Use approximately 5–7 image panels only.\n- Avoid a dense collage.\n\nEach panel:\n- Rounded rectangle.\n- Thin subtle border.\n- Premium glossy display appearance.\n- Slight perspective distortion following the curve.\n- Gradually decrease in size toward the background.\n- Use realistic generated landscape photography:\n    • alpine lake\n    • mountain valley\n    • waterfall\n    • dramatic mountain peaks\n    • coastal sunset\n- Rich but natural colors.\n- Foreground panels sharp, distant panels progressively softer / slightly blurred.\n- Strong depth-of-field.\n\nThe ribbon itself should have an extremely subtle luminous edge:\n- cyan / electric blue on the left\n- transitioning gently toward violet / magenta on the right\n\nKeep the glow restrained and premium.\nDo not make it look like cyberpunk neon.\n\nLIGHTING\n\nAdd one very thin horizontal blue-to-violet light flare passing subtly behind the “6-step” text.\n\nUse soft reflected blue/violet light underneath the image ribbon.\n\nLighting should feel cinematic and expensive, not game-like.\n\nBOTTOM-RIGHT\n\nSmall footer:\n“Generate with viggle-turbo”\n\n- Small clean sans-serif.\n- White / light gray.\n- Bottom-right aligned.\n- Approximately 5% from the right and 6–8% from the bottom.\n- Lots of breathing room.\n- Do not make it prominent.\n\nOVERALL COMPOSITION\n\nThe visual hierarchy should clearly be:\n\n1. “6-step generation”\n2. “Qwen-Image-2.1 ❤️ Viggle Turbo”\n3. flowing generated-image ribbon\n4. “DMD-distilled · Generate & Edit”\n5. small “Generate with viggle-turbo.” footer\n\nThe graphic should immediately communicate:\nfast generation, image generation, premium AI model, six-step inference.\n\nOverall aesthetic:\nminimal, elegant, expensive, technically sophisticated, cinematic, polished, restrained, premium AI launch campaign.\n\nAvoid:\n- excessive neon\n- cyberpunk look\n- too many thumbnails\n- busy collage layouts\n- random icons\n- fake UI elements\n- gradients inside every object\n- excessive lens flares\n- heavy shadows\n- 3D metallic typography\n- cartoon imagery\n- excessive text\n- logos other than the specified text\n- watermark\n- spelling errors\n- distorted letters\n- duplicated text",
   "steps": 8,
   "size_label": "3:2 · 2496×1664 (2048² area)",
@@ -112,7 +112,7 @@
    null
   ],
   "result": "qwen_perfume.png",
-  "info": "Pre-rendered example · seed `42` · 2496×1664 · 6 steps · v0.2.1 LoRA r256 · prompt sent as written (enhancement off)",
+  "info": "Pre-rendered example · seed `42` · 2496×1664 · 6 steps · v0.3 LoRA r256 · prompt sent as written (enhancement off)",
   "used_prompt": "Photograph a translucent emerald perfume bottle on pale limestone beside a shallow pool. Rippling sunlight reflects through the glass onto the stone. A single olive branch frames the upper left corner. Luxury product photography, realistic refraction, crisp bottle edges, soft shadows, uncluttered composition, no logo or text.",
   "steps": 6,
   "size_label": "3:2 · 2496×1664 (2048² area)",