22+ prompts. 223 things people built, tested and compared with Gemini 4 Carbon in its first weeks: 3D scenes, games, simulations, films made from code, real bug fixes and research, and dozens of same-prompt tests against Claude Fable 5.2 and Gemini 4 Argon. Each is credited and linked to the original post, with the prompt where the creator published one.
← All 223 Gemini 4 Carbon prompts
Last updated: · Gemini 4 Carbon API — pricing & playground · API docs
Gemini 4 Carbon (model id gemini-4-carbon) is Google's deeper-reasoning Gemini 4 model. Google built it on Gemini 4 Argon and describes the difference as more thinking tokens per turn; it opened to vetted cybersecurity defenders on 28 October 2026 and to every paid Gemini API developer on 10 November. It reads text, images, video, audio and PDFs with a 1,048,576-token context and writes text, so the scenes, games and films here are code it wrote (Three.js, WebGL, Blender Python, CUDA C++, Web Audio), mostly run by Google's Antigravity agent. Of the 223 cases, 18 include the prompt as published, 33 only the creator's request and 172 no prompt; we never fill one in. In same-prompt tests creators usually found Claude Fable 5.2's results more polished and Carbon's cost a small fraction of Fable's. On apimodels.app Carbon costs $1.10 per 1M input tokens and $5.50 per 1M output through 30 November 2026 and $2.20 / $11 from 1 December, against Google's $2 / $10 and $4 / $20.
Source: Google · Gemini 4 Carbon (the model vendor’s official site)
No prompt published. After two weeks of early access, the author made an entire music video with only code; more examples and the code are in the thread.
No prompt published. A single reference image turned into an animation made entirely of code: no skills or frameworks, just JavaScript and Python.
No prompt published. Starting from a static image and a text prompt, Carbon produced a 50-second animated Halloween walk with stereo audio.
Build a deterministic, seamless-looping 3-minute (180.000 s, 10,800 frames at 1920x1080, 60 fps) solo-violin concert film titled "NOCTURNE IN D MINOR // SOLO VIOLIN" set in a moonlit Parisian Haussmannian salon, using headless Blender (5.2 LTS+ with Cycles GPU baking) for 3D character, cloth, soft-body, luthiery, and a …
This is an excerpt. Read the full prompt in the original post →No prompt published. The author has been making music from everyday sounds with Carbon, with a prototype link and a how-it-was-made thread.
No prompt published. Carbon made a video with plain code; nothing in it comes from a video model, and every frame is math. The author keeps rewatching it.
No prompt published. The author wrote a song and turned it into a music video titled 追光末班车 using only Carbon and a cloud G4 GPU, and was surprised that Carbon could handle the arrangement and vocals.
Gemini 4 Carbonの性能の高さがわかる凄いMVを作成して！
Only the creator's request is public, not a full prompt: "Build a deterministic, real time capable 3 minute (180.000s, 10,800 frames at 60fps) solo piano concert film titled "PERF-01 // SYMPHONY NO. 7 IN A MAJOR, OP. 92 (II. …"
自以太坊首个价格日持有到今天，对比btc，你做一个酷炫的视频出来，不设限制
No prompt published. A short period-drama film titled 步步惊心一念三百年, made with Carbon controlling the pipeline together with Colab; it used 2% of the author's weekly quota.
请使用blender制作一个叙事性mv动画，风格偏向pv短片，背景音乐使用ref.mp3。2.动画主题是在明日方舟世界观，以干员玫兰莎视角看待战争与冲突。3.自由发挥，质量尽可能最好，4.不要吝啬token和工具调用，分镜连贯流畅，需要什么工具直接下，5。玫兰莎的参考设定图在文件夹里。可以用nanobanana来生图或者修改网上搜到的图，尽量风格化一下让ai图不那么出戏。6.可以使用gemini flow的omni 1.2来补充细节（仅在必要时） 7.你具有决策权，只是不要做成那种“ai味很重的，看起来一股廉价web感”的mv。一切质量优先。如果你不确定，可以先去看看何为ai味视频的讨论和案例 ，并以pv关键词在bilibili搜索视频学习何为优秀案例。
YouTubeNo prompt published. With the same prompt and 2,000 Higgsfield credits, each model made a short film on its own (Fable's "Keeper", Carbon's "LOCKOUT"), judged over five rounds: story, dialogue, continuity, look and working style.
YouTubeNo prompt published. Fable 5.2 and Carbon each described a 30-second dinosaur doodle and wrote the image and video prompts; the creator then made character sheets, a first frame and the film with GPT Image 2.6 and Seedance 2.5 on Higgsfield. Carbon took round one and Fable got one rematch.
No prompt published. Carbon made the pure-code short film 《刹那》 with about 1M tokens, as a response to a short film made with Fable 5.2. The author says the result is actually not as good as the original Fable-made film.
No prompt published. A side-by-side, code-generated stickman wuxia fight animation from Fable 5.2 and Carbon.
No prompt published. Carbon led a music video with a live audio waveform on the left and a one-take dance on the right. It follows an earlier Opus-made version, and the author asks viewers which is better.
No prompt published. Carbon (medium, the default), Fable 5.2 (xhigh) and QoDuck (extra high; the title calls it Qwen 4 Max Preview) got the same prompt for a 3D rocket launch animation. All three are recorded launching simultaneously for a blind A/B/C comparison.
/swarm 看一下这个视频：BV1z7nC6GENd，找出原始提示词存到目录下然后创作一个艺术品交付给我。
Make a three js animation where yonkers stands up as a big mech (cool stand up motion) and stomps around manhattan and queens and Brooklyn, use openstreetmap mapping data
No prompt published. Carbon was allowed to use every Antigravity tool, including Banana 2.1, to make a video from one sentence. The author found that adding the image model backfired, because the cut-outs were not clean.
Prompt formula
Carbon build brief = the deliverable and format (one HTML file, an MP4, a repo) + the stack and constraints + what done looks like + a reference (an image or video it should match) + a self-check (open it in a browser, screenshot, fix) + a stop condition or time and token cap
Example (our template, not a creator's prompt): Build a browser game: a night-time arcade racer on a neon highway. Three.js from a CDN, music and sound in Web Audio, everything else in one index.html. Match the mood of the attached video. Done means: it loads with no console errors, steering and crashes work, and a lap takes about 90 seconds. Open it in a headless browser, take screenshots at the start and mid-lap, fix what looks wrong, then tell me what you checked. Do not ask questions.
Not directly. Carbon takes text, images, video, audio and PDFs and returns text. Every scene, game and film here was rendered from code it wrote: Three.js and WebGL pages, Blender Python, CUDA or C++ programs, Web Audio for sound. In a chat window you get the code or an HTML file; an agent such as Antigravity, Cursor or GitHub Copilot can run, render and test it.
Mostly Google's Antigravity, often with its /swarm multi-agent mode. Others used Cursor, GitHub Copilot, AI Studio, Vertex AI or the API from their own harness. Where the creator said, the card names the tool.
Usually not on polish, often yes on cost. In most same-prompt tests here Fable 5.2 produced the better-looking, more complete result, while Carbon cost a fraction as much and sometimes won outright: $7.54 against $41.71 for three jobs in one test, where Carbon won two of the three.
Mostly yes, at the cost of time and tokens. The two cost the same per token; in these tests Carbon usually produced the stronger result and took several times longer, for example 81 minutes against 14 in one Blender test that Argon had failed. Google still recommends Argon for most everyday coding.
As the creators reported: $0.45 for a one-shot 3D platformer of 2,599 lines, $3.82 for the shooter rebuild, about $7.40 for four rounds of self-review and $27.68 for three long builds. Cost depends on the thinking level and how long the agent runs; thinking tokens are billed as output.
On apimodels.app, call model gemini-4-carbon on the OpenAI-compatible /v1/chat/completions endpoint or the native Gemini endpoint with one key, from anywhere including mainland China. Raise the output limit to at least 32,768 tokens, stream long answers and allow a 15-minute timeout, because Carbon thinks for a while before it writes.
The works belong to their creators. Clips from X play from X's own servers, videos on YouTube and Bilibili play in their own players, and every card links to the original; we do not host or re-upload any of them. Published prompts are quoted with credit, in full when short and excerpted when long. If you made one of these and want the card removed or the credit corrected, email us.
Gemini 4 Carbon is live here alongside 160+ other image, video, audio and language models — one API, one key, pay as you go, with $0.10 of trial credit on sign-up.
The whole collection is on GitHub as Markdown and JSON: all 223 cases with credits and links to the original posts and pages, the tools and numbers each creator reported, and a summary of the same-prompt tests. The videos and images stay with their creators and are linked, not re-hosted. Prompts up to 400 characters are quoted in full; longer ones are excerpted, and unpublished ones are left out. Our own write-ups are licensed CC BY 4.0. Open an issue to add a case or correct a credit.
View on GitHub