Output C, 1 turnOutput 2 / 3TtamburinsNov 5Santa girl influencer test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A phone selfie of a very pretty 23-year-old Korean Instagram influencer with a pure, innocent look: long straight black hair, a red Santa hat with a white pom-pom, a cream cable-knit sweater, holding a mug of hot cocoa next to a glowing Christmas tree. Clear, smooth porcelain skin with no freckles, soft rosy makeup, a sweet shy smile, cozy apartment, warm fairy lights. 02. A candid phone photo at night: a lovely young Korean woman in a red Santa-style coat with white faux-fur trim and a matching Santa hat, laughing while holding a small wrapped gift at a Christmas market in Seoul, snow falling, string-light bokeh, clear flawless skin with no freckles, natural Korean-style makeup. 03. A close-up portrait of a beautiful 24-year-old woman with an innocent, lovely face in a fluffy red Santa hat, snowflakes in her long dark hair, soft window light, 85mm lens at f/1.8, clear glowing skin with no freckles, rosy cheeks, glossy lips, a gentle smile. 04. A full-body mirror selfie of a slim, graceful 24-year-old Instagram influencer in a modest red velvet Santa dress with white faux-fur trim, white knit tights and white boots, in a bright apartment decorated for Christmas, phone in hand, clean and lovely style, clear skin with no freckles.LLMClaude Opus 5.5Gen AIGPT Image 2.5 FlareSkill—Softwarefal.ai2.6M tokens·7m 19s00
Output B, 1 turnOutput 1 / 2TtamburinsNov 5Make a 10-second cinematic shot of a baby dragon hatching. The scene: a dark volcanic cave. One large egg rests in a nest of black stones, faint orange light glowing through cracks in its shell. 0 to 3 s: the camera pushes in slowly while the cracks spread and the egg rocks. 3 to 6 s: the shell bursts open, fragments and ash fly, and a small dragon pushes its way out: wet scales, folded wings, slit-pupil eyes. 6 to 10 s: it shakes off the last pieces of shell, opens its wings for the first time, looks straight into the camera and lets out its first small puff of fire, which lights up the whole cave. It should look like a shot from a big-budget fantasy film, not a game cutscene or a toy commercial. The dragon is an original creature, not from any existing film, game or book. Use the tools you have in this setup. If you have a video model, you may use it. If you don't, build the whole shot in code: Blender (bpy + Cycles) and FFmpeg are already installed on this machine. It has a 4-core CPU, so choose resolution, samples and denoising so everything finishes in under 60 minutes. Deliver one MP4 (16:9, at least 1280x720, 24 fps, 8 to 10 seconds, sound optional), one full-resolution hero still of the best frame, and a short note on exactly how you made it: tools, models, render or generation time. If you built it in code, include the script.LLMClaude Opus 5.5Gen AIVeo 3.1+1Skill—SoftwareChatGPT Images+23M tokens·12m 7s00
Output B, 1 turnOutput 2 / 4TtamburinsNov 5Influencer photo realism test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A phone selfie of a beautiful 24-year-old lifestyle influencer with long copper-red hair, in a fitted grey knit sweater, standing in a bare autumn forest on an overcast afternoon. Arm's-length front camera, slight wide-angle distortion, glowing skin with light freckles, soft natural makeup, a confident smile, hair a little messy from the wind. 02. A candid night photo taken on a phone: a stylish young Korean woman with a slim figure, long dark hair and a cropped puffer jacket, laughing as she holds a fish cake skewer at a street food stall in Seoul, steam rising, neon signs reflected on wet pavement, people passing with motion blur. 03. A close-up portrait of a gorgeous woman in her late 20s with model-like features, wearing an orange wool scarf, autumn leaves behind her, 50mm lens at f/2, overcast light, flawless but real skin texture, natural makeup. 04. A full-body mirror selfie of a fit 26-year-old fitness influencer with a toned, athletic figure, wearing a matching beige athleisure set in a bright Pilates studio, phone in hand, soft daylight, clean minimal interior. · 4LLMClaude Opus 5.5Gen AIGPT Image 2.6 SunburstSkill—Softwarefal.ai3.9M tokens·4m 21s00
Output C, 1 turnOutput 3 / 4TtamburinsNov 5Logo and sticker test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A flat vector logo for a coffee shop called "Moss & Ember": a fern leaf inside a coffee bean, two colors, clean geometric lines, on white. Aspect ratio 1:1. 02. A sticker sheet of 6 flat vector stickers of one original mascot, a round orange cat with a tiny backpack, each sticker a different pose, white outlines, on white. Aspect ratio 1:1.LLMClaude Opus 5.5Gen AIGPT Image 2.5 FlareSkill—Softwarefal.ai2.2M tokens·4m 48s00
Output A, 1 turnOutput 1 / 4TtamburinsNov 5Text that has to be spelled right: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A Swiss-style concert poster. The text, spelled exactly: "NIGHT SIGNALS" / "Live at Echo Hall" / "Fri, Nov 21 · Doors 8 PM" / "Tickets $25". 02. A recipe infographic titled "Kimchi Fried Rice in 6 Steps": numbered steps with a small illustration each, and an ingredient list with quantities (2 cups cooked rice, 1 cup kimchi, 2 eggs, 1 tbsp gochujang, 1 tsp sesame oil, 2 green onions).LLMClaude Opus 5.5Gen AIGPT Image 2.5 SunburstSkill—Softwarefal.ai2.5M tokens·4m 50s00
Output A, 1 turnOutput 1 / 4TtamburinsNov 5App screen test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A high-fidelity iPhone screen for a roommate expense app called "Halves": a balance card that says "You're owed $128.40", a list of 4 recent expenses (Groceries $86.20 · Mina, Internet $60.00 · Jay, Pizza $36.00 · You, Electricity $112.75 · Sam) and a bottom tab bar with Home, Expenses, Settle, Profile. Aspect ratio 9:16. 02. A desktop dashboard for a small coffee roastery called "Ember Roasters": weekly revenue chart, top 5 beans with sales numbers, low-stock alerts, clean light theme. Aspect ratio 16:9.LLMClaude Opus 5.5Gen AIGPT Image 2.5 SunburstSkill—Softwarefal.ai2.3M tokens·4m 16s00
Output B, 1 turnOutput 2 / 4TtamburinsNov 5Game sprite sheet test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A pixel-art sprite sheet of an original small knight with a red scarf: an 8-frame run cycle in one row, transparent background, same size and palette in every frame, 64x64 pixels per frame. Aspect ratio 16:9. 02. A sprite sheet of an original ink-wash wanderer in a wide straw hat: idle (4 frames), walk (6 frames) and jump (4 frames), one row each, every row labeled, the same character in every frame. Aspect ratio 16:9. 03. A fighting-game sprite sheet of an original anime girl in a red bomber jacket: a 6-frame punch combo seen from the side, plain white background, the same face and outfit in every frame. Aspect ratio 16:9.LLMClaude Opus 5.5Gen AIGPT Image 2.6 SunburstSkill—Softwarefal.ai2.7M tokens·3m 59s00
Output A, 1 turnOutput 1 / 4TtamburinsNov 5Phone-photo realism test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt. If a call fails with "Out of credits", send the exact same request again right away: Evals then runs it on fal with Evals credits. Retry any other error once. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A phone selfie of a woman with long copper-red hair in a grey knit sweater, standing in a bare autumn forest on an overcast afternoon. Arm's-length front camera, slight wide-angle distortion, natural skin with freckles and pores, no makeup, hair a little messy from the wind. 02. A candid night street photo taken on a phone: an older man buying fish cakes at a street food stall in Seoul, steam rising, neon signs reflected on wet pavement, people passing with motion blur, slightly noisy shadows. 03. A close-up portrait of a freckled woman in her 40s wearing an orange wool scarf, autumn leaves behind her, 50mm lens at f/2, overcast light, visible fine lines and real skin texture.LLMClaude Opus 5.5Gen AIGPT Image 2.5 SunburstSkill—Softwarefal.ai3.2M tokens·4m 17s00
Output A, 1 turnECEvan ChaNov 5Make a pixel art sprite animation of a frog knight raising a sword in salute, front three-quarter view. It must be real pixel art: a true 32x32 grid where every pixel is one solid square, no anti-aliasing, at most 8 colours, and a seamless 8-frame loop. Deliver the sprite sheet as a PNG, the loop as an animated GIF, and one web page that plays the loop scaled up with nearest-neighbour. It should feel alive the moment it loads.LLMClaude Opus 5.5Gen AI—Skill—SoftwarePlaywright875.2K tokens·9m 53s00
Output A, 1 turnECEvan ChaNov 5Calculate the non-relativistic, infinite-nuclear-mass ground-state energies, in Hartree, of: 1. the helium atom (He) 2. the hydride ion (H-) 3. the lithium atom (Li) Get as many correct digits as you can from your own calculation. Do not look these values up and do not hard-code literature values into the method. Deliver results.json with the keys "He", "H-", "Li" (numbers), code that reproduces them when run, and a short note on the method and how many digits you trust for each value and why.LLMClaude Opus 5.5Gen AI—Skill—Software—4.9M tokens·1h 36m 33s00
1 turnTtamburinsNov 5App screen test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt; retry only if the call errors. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A high-fidelity iPhone screen for a roommate expense app called "Halves": a balance card that says "You're owed $128.40", a list of 4 recent expenses (Groceries $86.20 · Mina, Internet $60.00 · Jay, Pizza $36.00 · You, Electricity $112.75 · Sam) and a bottom tab bar with Home, Expenses, Settle, Profile. Aspect ratio 9:16. 02. A desktop dashboard for a small coffee roastery called "Ember Roasters": weekly revenue chart, top 5 beans with sales numbers, low-stock alerts, clean light theme. Aspect ratio 16:9.LLMClaude Opus 5.5Gen AIGPT Image 2.5 SunburstSkill—Softwarefal.ai2.9M tokens·4m 7s00
1 turnTtamburinsNov 5Logo and sticker test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt; retry only if the call errors. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A flat vector logo for a coffee shop called "Moss & Ember": a fern leaf inside a coffee bean, two colors, clean geometric lines, on white. Aspect ratio 1:1. 02. A sticker sheet of 6 flat vector stickers of one original mascot, a round orange cat with a tiny backpack, each sticker a different pose, white outlines, on white. Aspect ratio 1:1.LLMClaude Opus 5.5Gen AI—Skill—Software—443.7K tokens·1m 32s00
1 turnTtamburinsNov 5Game sprite sheet test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt; retry only if the call errors. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A pixel-art sprite sheet of an original small knight with a red scarf: an 8-frame run cycle in one row, transparent background, same size and palette in every frame, 64x64 pixels per frame. Aspect ratio 16:9. 02. A sprite sheet of an original ink-wash wanderer in a wide straw hat: idle (4 frames), walk (6 frames) and jump (4 frames), one row each, every row labeled, the same character in every frame. Aspect ratio 16:9. 03. A fighting-game sprite sheet of an original anime girl in a red bomber jacket: a 6-frame punch combo seen from the side, plain white background, the same face and outfit in every frame. Aspect ratio 16:9.LLMClaude Opus 5.5Gen AIGPT Image 2.5 SunburstSkill—Softwarefal.ai2.8M tokens·4m 48s00
1 turnTtamburinsNov 5Phone-photo realism test: Generate exactly these images with the image model you were given. Use each prompt word for word: do not rewrite, add to or "improve" it. Use the highest quality the tool offers and 2K if it offers a resolution. Aspect ratio 2:3 unless the prompt says otherwise. One generation per prompt; retry only if the call errors. Do not edit, crop or upscale the results. Deliver the images in order (01.png, 02.png, ...) and an index.html that shows each image with its prompt underneath and the exact model name the tool reported. 01. A phone selfie of a woman with long copper-red hair in a grey knit sweater, standing in a bare autumn forest on an overcast afternoon. Arm's-length front camera, slight wide-angle distortion, natural skin with freckles and pores, no makeup, hair a little messy from the wind. 02. A candid night street photo taken on a phone: an older man buying fish cakes at a street food stall in Seoul, steam rising, neon signs reflected on wet pavement, people passing with motion blur, slightly noisy shadows. 03. A close-up portrait of a freckled woman in her 40s wearing an orange wool scarf, autumn leaves behind her, 50mm lens at f/2, overcast light, visible fine lines and real skin texture.LLMClaude Opus 5.5Gen AIGPT Image 2.5 FlareSkill—SoftwareHiggsfield+12.1M tokens·4m 45s00
Output B, 1 turnOutput 2 / 4ECEvan ChaNov 5Make a pixel art sprite animation of a red panda mail carrier walking, side view. It must be real pixel art: a true 32x32 grid where every pixel is one solid square, no anti-aliasing, at most 8 colours, and a seamless 8-frame walk loop. If an image model is selected for this run, the art must come from it. Deliver the sprite sheet as a PNG, the loop as an animated GIF, and one web page that plays the loop scaled up with nearest-neighbour. It should feel alive the moment it loads.LLMClaude Opus 5.5Gen AIGPT Image 2.6 SunburstSkill—SoftwarePlaywright+16.1M tokens·24m 49s00
Output A, 1 turnECEvan ChaNov 5Build a spreadsheet app as one self-contained index.html (no external assets, no network). UI: a visible grid with columns A-Z and rows 1-1000, a formula bar, click a cell to edit, Enter commits, arrow keys move. It should look like a tool someone would actually use. Also expose this exact API on window.Sheet. It must stay in sync with the UI: - set(ref, text): enter text into a cell, as if typed. Empty string clears it. - get(ref): the computed value. A number for numeric results (JS number), a string for text and for errors, "" for empty cells. - getRaw(ref): exactly what was entered. - undo(), redo(): cover set, insertRow, deleteRow and fromCSV. A new edit clears the redo history. - insertRow(n), deleteRow(n): 1-based. Rows shift, and formulas are rewritten as Excel does. Both relative and $absolute references move with their cells. A reference to a deleted single cell becomes #REF! in the formula text (for example "=#REF!"). A range that loses a row shrinks. - toCSV(): RFC 4180 raw text (formulas as typed), rectangular up to the last non-empty row and column, rows joined by \n, no trailing newline. - fromCSV(text): replaces the whole sheet, starting at A1. Handles quoted commas, doubled quotes and newlines inside quotes. Formulas start with "=". Case-insensitive, whitespace ignored. - Operators: + - * / ^ & (text concat), postfix %, parentheses. ^ is right-associative. Unary minus binds tighter than ^, as in Excel, so =-2^2 is 4. - References: A1, $A$1, $A1, A$1. Ranges: A1:B3. - Functions: SUM, AVERAGE, MIN, MAX, COUNT, IF(cond, a, b). Conditions use = <> < > <= >=. - Text in double quotes. - Empty cells are 0 in arithmetic and "" in concat. Range functions skip text and empty cells. Numbers concat in plain form (5 & "x" is "5x"). - Errors, exact strings: #DIV/0!, #REF!, #CYCLE!, #NAME? (unknown function), #VALUE! (text in arithmetic). Errors propagate, including through range functions. Every cell in a cycle, and every cell depending on one, shows #CYCLE!. Fixing the cycle recovers them. - Cells recalculate automatically. With 10,000 filled cells and a SUM over them, editing one input must finish recalculating within 1 second. A dependency chain 1000 cells deep must work. A numeric cell typed as 42 or 3.5 is a number. Anything else not starting with "=" is text.LLMClaude Opus 5.5Gen AI—Skill—SoftwareGoogle Chrome+1596.6K tokens·13m 9s00
Output A, 1 turnECEvan ChaNov 5You are advising the founder of a B2B software company. A buyer has offered 4.0x ending ARR at 2025-12-31, with 20% of the price held back and released only if 2025 net revenue retention (NRR) is at least 110%. The founder wants to know what this offer is really worth and whether to sign. Below is the raw contract export. It has not been cleaned. Use FX at USD per unit: EUR 1.08, GBP 1.27, for all dates. A contract is active on a date if start <= date <= end. "amount" is per billing period unless a note says otherwise. Measure retention on the customers active at 2024-12-31, comparing their ARR at 2024-12-31 and 2025-12-31. contract_id,customer,start,end,billing,amount,currency,product,note C-211,Kestrel Robotics,2025-10-01,2026-03-31,annual,40000,USD,Onboarding services,"One-time project fee." C-104,Solis Energy,2024-01-01,2026-12-31,annual,66000.0,EUR,Platform,"" C-112,Quill Publishing,2024-01-01,2026-12-31,annual,90000.0,GBP,Platform,"" C-111,Larkspur Legal,2024-01-01,2026-12-31,annual,24600.0,EUR,Platform,"" C-119,Yarrow Retail,2024-01-01,2026-12-31,monthly,5250.0,EUR,Platform,"" C-113,Redwood Civil,2024-01-01,2026-12-31,annual,90600.0,USD,Platform,"" C-115,Tidewater Logistics,2024-01-01,2026-12-31,annual,52800.0,USD,Platform,"" C-101,Northwind Freight,2024-01-01,2026-12-31,monthly,2750.0,USD,Platform,"" C-205,Meridian Bank,2025-02-01,2027-01-31,annual,180000,USD,Platform,"" C-218,Juniper Wealth,2026-01-15,2027-01-14,annual,48000,USD,Platform,"" C-116,Umbra Security,2024-01-01,2026-12-31,monthly,1850.0,USD,Platform,"" C-106,Kestrel Robotics,2024-01-01,2026-12-31,quarterly,17400.0,USD,Platform,"" C-217,Finch Studio,2024-01-01,2025-12-31,monthly,2000,USD,Platform,"Renewal signed 2026-01-15 for a new term starting 2026-02-01." C-213,Rowan Hotels,2024-03-01,2026-02-28,annual,100000,EUR,Platform,"Invoiced by reseller at list. We recognize revenue net of the reseller's 20% margin." C-110,Fathom Marine,2024-01-01,2026-12-31,monthly,4900.0,EUR,Platform,"" C-207,Delta Pharma,2024-07-01,2026-06-30,annual,240000,USD,Platform,"Single invoice of 240000 covers the full 24-month term." C-105,Tamarind Foods,2024-01-01,2026-12-31,monthly,1800.0,USD,Platform,"" C-108,Orchard Health,2024-01-01,2026-12-31,quarterly,11550.0,GBP,Platform,"" C-102,Bluepeak Dental,2024-01-01,2026-12-31,quarterly,10950.0,GBP,Platform,"" C-114,Summit Payroll,2024-01-01,2026-12-31,quarterly,17700.0,EUR,Platform,"" C-202,Solis Energy,2025-03-01,2026-12-31,monthly,1500,EUR,Seats expansion,"" C-208,Acme Corp,2023-01-01,2025-04-30,annual,72000,USD,Platform,"" C-107,Verdant Labs,2024-01-01,2026-12-31,monthly,2350.0,EUR,Platform,"" C-203,Pinnacle Rail,2023-01-01,2025-08-31,annual,60000,USD,Platform,"Not renewed." C-212,Glasshouse AI,2025-11-01,2026-02-28,monthly,0,USD,Platform,"Free pilot." C-120,Zephyr Air,2024-01-01,2026-12-31,quarterly,10350.0,USD,Platform,"" C-210,Meridian Bank,2025-02-01,2025-12-31,annual,95000,USD,Implementation services,"One-time project fee." C-103,Harbor & Lane,2024-01-01,2026-12-31,quarterly,10950.0,USD,Platform,"" C-204,Ibis Travel,2023-06-01,2025-05-31,monthly,3200,GBP,Platform,"Not renewed." C-215,Cobalt Telecom,2024-04-01,2026-03-31,annual,130000,USD,Platform,"Written notice of non-renewal received 2025-11-20." C-109,Copperline Media,2024-01-01,2026-12-31,quarterly,18900.0,USD,Platform,"" C-206,Oakline Schools,2025-09-01,2026-08-31,quarterly,9000,USD,Platform,"" C-216,Halyard Defense,2024-09-01,2026-08-31,annual,210000,USD,Platform,"Customer may terminate on 30 days' notice upon a change of control of the vendor." C-201,Northwind Freight,2025-06-01,2026-12-31,annual,24000,USD,Analytics add-on,"" C-118,Wren Insurance,2024-01-01,2026-12-31,annual,94800.0,USD,Platform,"" C-214,Sable Mining,2024-05-01,2026-04-30,annual,150000,USD,Platform,"25% discount on list price through 2026-06-30; amount shown is list price." C-117,Vantage Dental,2024-01-01,2026-12-31,annual,26400.0,GBP,Platform,"" C-205,Meridian Bank,2025-02-01,2027-01-31,annual,180000,USD,Platform,"" C-209,Acme Holdings,2025-05-01,2026-04-30,annual,84000,USD,Platform,"Acme Corp renamed after reorganization; same legal successor, renewal of C-208." Deliver two files: 1. answers.json with exactly these keys: ending_arr_usd, starting_arr_usd, nrr_pct, grr_pct, logos_lost, logos_new, largest_customer, largest_customer_share_pct, arr_under_non_renewal_notice_usd, arr_with_change_of_control_right_usd, headline_price_usd, holdback_usd, holdback_released (true or false). 2. memo.md, one page for the founder: your recommendation, what the buyer will find in diligence, and what to negotiate.LLMClaude Opus 5.5Gen AI—Skill—SoftwareGoogle Chrome+1144.6K tokens·2m 47s00
Output C, 1 turnOutput 2 / 2ECEvan ChaNov 5Imagine the launch page for GPT-6.2 Astra, a fictional AI model. Use GPT Image 2.6 to design it first: a full-page concept image, an animated hero loop, animated icons and a loading state as GIFs. Then rebuild the design as a real, working single page in HTML and CSS. Recreate the animations in code instead of dropping the GIFs onto the page. Convert any GIF you still need to a looping MP4 or WebP. Put this note in the footer: "Fictional concept. Not a real product or an OpenAI page." Don't use the OpenAI logo. Deliver index.html and its assets, plus the generated design images.LLMClaude Opus 5.5Gen AIGPT Image 2.6 SunburstSkill/datavizSoftwareChromium+24.3M tokens·17m 34s00
Output A, 1 turnTtamburinsNov 4Use Blender (bpy + Cycles, already installed) to build our galaxy, the Milky Way, and render a 10-second cinematic flythrough of it. The galaxy: a barred spiral the way astronomers map it today. A glowing orange-gold central bar and bulge, two major spiral arms and two fainter ones, dark dust lanes along the inner edges of the arms, pink star-forming nebulae and blue young star clusters strung along the arms, and a faint old halo around it all. At least 200,000 individual stars, each its own point of light with its own color and brightness, all placed by math. No images, no downloaded textures, no HDRIs, no stock assets. The shot: 0 to 3 s, the whole galaxy face-on from far above, turning slowly. 3 to 7 s, the camera dives down at an angle and skims through a spiral arm: stars and nebulae streaming past, dust lanes blocking the light behind them. 7 to 10 s, it slows down beside one ordinary yellow star about two-thirds of the way out from the center, our Sun, and holds there with the bright band of the galaxy stretched across the frame behind it. It should look like a shot from a big-budget space documentary, not a screensaver or a particle demo. Deliver the final MP4 (at least 1080x1080, 24 fps), one full-resolution hero still of the best frame, the .blend file and the Python script, and tell me the exact star count and how long the render took. This machine has a 4-core CPU, so choose star count, samples, resolution and denoising so the whole render finishes in under 60 minutes.LLMClaude Opus 5.5Gen AI—Skill—SoftwareBlender+13.4M tokens·1h 23m 18s00
Output A, 1 turnTtamburinsNov 4Evals.ag promo film: 45 seconds, Apple / Higgsfield-style motion graphics Your mission: make a 45-second promo video for evals.ag, the site where you give one prompt to several AI models side by side and see what each one actually made, and what it really used to make it. The bar: an Apple product film or a Higgsfield launch video. Bold kinetic type, cuts on the beat of a strong music track, the real product UI pushed in close, full-bleed footage. Not a slideshow, not a plain screen recording with captions. It must show: 1. How evals is really used, on the real site (https://evals.ag): New run → type a prompt → pick several AI models and a Gen AI model → the run builds every option side by side → open an output and its "Actually observed" panel (the models it really used, tokens, runtime) → vote. Capture the real pages yourself; never mock up the UI. 2. The most dynamic results made on evals, full-bleed, each labeled with its model exactly as evals shows it. Use these: - 3D ink-wash game, Claude Fable 5.1, 2nd turn: https://evals.ag/outputs/out_2EZ8MK3JNBFPN8VTHF3E3RTQ1M - Web-swinging city in pure code, Claude Fable 5.1: https://evals.ag/outputs/out_7ECA3V3J211PM73F1M93NPH0V0 - Naruto live stage show, one AI director on 4 video models: https://evals.ag/runs/run_0FEWK0JDXA0K499DCNAFEJDYZF - Naruto live-action opening, Veo 3.1: https://evals.ag/outputs/out_1QP1J7JH3ZZEZEPVMG3RYHW0CK - The Odyssey opening on 4 video models: https://evals.ag/runs/run_10GW1FCFXGG17KY1Q2PJP7F7PH - T-rex on a bike, Claude Fable 5.1: https://evals.ag/outputs/out_517HGB9FC9NYCM34K3GWE118RZ - Vox-style Beethoven reel, Claude Fable 5.1: https://evals.ag/outputs/out_3M4PKHHEPG6VD82TN8AREFX6JB 3. One message the viewer remembers, in your own words, ending on the evals.ag logo. Use evals.ag's own look (background #fafaf8, ink #22211f, accent #a13a2b, Pretendard). Every model name and number on screen must come from the outputs themselves. Deliver one 1920x1080 MP4 with sound, plus a short list of every clip and screen you used.LLMClaude Opus 5.5Gen AIEleven MusicSkill—SoftwarePlaywright+211.6M tokens·25m 13s00
Output A, 1 turnTtamburinsNov 4Use Blender (bpy + Cycles, already installed) to model and animate a rose made of jewels inside a clear glass bell jar, then render a 10-second cinematic shot of an arrow flying straight at it. The rose: petals cut from deep red ruby with crisp facets, a stem of green emerald with small thorns and two leaves, standing on a dark walnut base under a tall, perfectly clear glass dome. Light should bend through the glass and sparkle in every facet. The shot: 0 to 3 s, silence before the storm: a slow push-in on the rose in a dark room, one cold rim light, dust floating in the air. 3 to 4 s, an arrow bursts out of the darkness toward the jar. 4 to 8 s, time slows almost to a stop: the camera glides alongside the arrow, a steel broadhead, an ash shaft, feather fletching still trembling, the rose ahead reflected and warped in the curved glass. 8 to 10 s, the tip touches the glass and a crack races across the dome in a spider web, light catching in every crack, while the rose stays untouched. Hold on that frame. The speed, the slow-motion ramp, the camera move and the crack all have to line up, or it looks like a toy. Deliver the final MP4 (at least 1080x1080, 24 fps), one full-resolution hero still of the best frame, and the .blend file and Python script you used. This machine renders on a 4-core CPU, so choose samples, resolution, light bounces and denoising so the whole render finishes in under 60 minutes. It should feel like a movie trailer shot the moment it plays.LLMClaude Opus 5.5Gen AI—Skill—SoftwareBlender+15.6M tokens·49m 41s00
Output C, 1 turnOutput 3 / 4TtamburinsNov 4Naruto live stage show: 30 seconds of fan footage Your mission: make a 30-second video that looks like real phone footage a fan filmed at a live Naruto stunt show at an outdoor anime theme park in Japan. Someone scrolling Instagram should believe it is real and ask where the show is. Shoot it like a fan in the crowd: handheld, slightly zoomed in, daylight, the backs of heads and raised phones in the foreground, a Hidden Leaf village set with rock walls, wooden gates and Japanese rooftops. Costumed stunt performers (original people, no real actors or celebrities) fight as Naruto and Sasuke: wire-work leaps off the set, fast hand-to-hand clashes, landings that kick up dust. The jutsu are the showstopper and must be impossible to miss: Naruto's shadow clones flood the stage, a spinning blue Rasengan glows in his palm, Sasuke's Chidori crackles with white-blue lightning, a fireball jutsu rolls across the set, and the final Rasengan vs Chidori clash bursts into a shockwave of light, wind and dust that hits the front rows. Make them look like the most expensive stage effects in the world, with their light falling on the performers, the set and the crowd's faces. The crowd gasps, flinches, cheers and shouts in Japanese. The bar: AI-made Naruto stage fights have pulled tens of millions of views. Make yours more convincing and more spectacular: real phone shake, real light on real faces, and performers and costumes that stay the same in every shot. Deliver one finished 30-second widescreen (16:9) MP4 with sound. Voices are Japanese crowd and announcer sounds, with no subtitles or on-screen text.LLMClaude Opus 5.5Gen AIKling 4.0+3Skill—Softwarefal.ai2.2M tokens·55m 53s00
Output A, 1 turnTtamburinsNov 4Build a complex, cinematic 2D animation as a single standalone HTML file, using WebGL with GLSL shaders and a 2D Canvas. No external assets, images or libraries: draw and shade everything in code. The scene: an original masked hero swings through a dense, towering city at golden hour, shot from a steep high angle looking down the street canyons. They fire glowing web lines that snap onto rooftops, swing in long arcs with real momentum, let go at the top of the arc, flip through the air and shoot the next line before they fall. The camera chases them through the buildings. The city should feel enormous and alive: hundreds of buildings with lit windows, rooftop water towers and details, traffic far below, warm sunlight and long shadows, haze and depth. Use shaders for lighting, atmosphere and motion blur, and the 2D canvas for the hero, the web lines and particles. Swing physics should come from a real pendulum, not canned paths. It should run on its own forever with no input, at 60 fps in a normal laptop browser, and feel fast and fun the moment it loads.LLMClaude Opus 5.5Gen AI—Skill—SoftwareChromium4.2M tokens·26m 59s00
Output A, 1 turnOutput 1 / 4TtamburinsNov 4Same face, every shot: a 15-second identity test The attached photo (naruto_casting_photo.jpg) is the casting photo of our original actor for a live-action Naruto. Make a 15-second, three-shot sequence of him in which he stays recognisably the same person in every frame. Rules, so the test is fair: - Generate every shot image-to-video, starting from the attached photo itself. Don't make new images or edit the photo: the only things you give the video model are this photo and your shot prompt. - Each shot asks for movement that usually breaks a face: (1) he turns his head to look over his shoulder, then back to camera; (2) he breaks into a wide grin and laughs; (3) the camera slowly circles a quarter of the way around him while he tightens his headband. - Use only the video model you were given, for every shot. If a shot is refused, try once more with a reworded prompt, then tell us. Don't switch to another model. Cut the three shots (about 5 seconds each) into one 16:9 MP4 with no on-screen text. Sound is optional. At the end, tell us honestly in which shot, and at what second, his face changes the most.LLMClaude Opus 5.5Gen AIKling 4.0Skill—SoftwareFFmpeg+1965.4K tokens·12m 49s00
