Optional and open to every team, in every track. You win it with the project you're already building, and entering takes about 30 minutes at the end.
| What you send | Two links, your 3 slides (from the template in this folder) and your 3 to 5 minute screen recording, plus your team name and everyone's emails |
| Where | The entry form (also pinned in Discord) |
| When | By Sunday 9:00 AM, the same deadline as your pitch deck and code |
| Prize | $200 in gift cards split across the winning team, plus qBraid credits for each member. Announced at the closing ceremony |
| Judged on | Did it run on qBraid, and how creatively did you use it? |
Use your credits. They're there to be used, not to sit in your wallet. Run on real hardware, launch a GPU, try qBraid AI, and explore everything in the qBraid docs. Two rules: do your track's hardware run first, and make every run do something for your project. GPU instances bill by the hour, so stop them when you're done. Benchmarking a device just to fill a table doesn't count.
Try a GPU. Once your problem grows past a few dozen qubits or a big optimization, a laptop-sized simulator can't keep up. Launch an on-demand GPU instance, or send circuits to qBraid's GPU simulators (how). A GPU run that moves your project forward counts in your favor.
-
Use qBraid in your project. You probably already are: if your notebook runs in qBraid Lab and sends a job to a qBraid device or simulator, you qualify. Want to stand out? Make it visual (about 5 minutes), or take your problem to a GPU.
-
Fill in the slides (10 minutes). The template is in this folder:
qbraid-award-template.pptx.- Get it: in qBraid Lab, right-click the file and choose Download.
- Open it: PowerPoint, Keynote or Google Slides all work. For Google Slides, upload it to Drive, then open it with Google Slides.
- Fill it in: every slide tells you what to write. The speaker notes tell you what to say in your video.
-
Record your screen (10 minutes). Walk through your slides, then show qBraid running. Use the script below.
-
Fill in the entry form (2 minutes). It's also pinned in Discord. One person enters for the whole team. The form asks for:
- your team name
- every team member's email (the one each person uses on qBraid), so the prize and credits reach everyone
- a link to your slides: Google Slides, or the file on Google Drive
- a link to your video: YouTube (unlisted), Vimeo, Loom or Google Drive
Check the sharing on both before you submit. Set each to "anyone with the link can view", then open it in a private browser window to make sure it plays or opens. If we can't open it, we can't judge it.
Slide 1 has your team name, track and members (names only; emails go in the form). If you made a qBraid artifact, put its link on your slides.
Judges love a visual, and this gets you one fast. Run this in a notebook cell after you have results:
import sys; sys.path.append("../4-qbraid-award")
from make_visual import make_visual
make_visual({"Simulator": sim_counts, "IonQ": hw_counts}, title="Our results") # or just one counts dictUse the names your notebook already has: plant scheduling make_visual({"Simulator": sim_counts, "IonQ": counts}); sensor placement make_visual({"IonQ": counts}); Grover make_visual({"Simulator": counts, "IonQ": hw_counts}).
-
The cell writes
visual.html, an animated bar chart, in the same folder as your notebook. -
Open it next to your notebook in qBraid Lab's Agent Canvas panel. Run this from a Lab terminal in that folder:
qbraid-canvas visual.html --title "Our results" -
Publish it as a qBraid artifact, so judges can open it from a link. From the same terminal:
qbraid artifacts publish visual.html --title "Our results" --visibility publicIt prints a link anyone can open. Put that link on your slides. If
qbraid artifactsisn't available in your terminal, ask a qBraid mentor. -
Make it yours. Ask your coding agent to restyle it, add controls, animate your algorithm step by step, or turn it into a game. To publish the new version to the same link, run
qbraid artifacts listfor its ID, thenqbraid artifacts update <ID> --file visual.html.
For inspiration, see Quantum Runner on qBraid's Explore page. It's a side-scroller where a real quantum measurement decides where the gaps go.
One file, three slides: qbraid-award-template.pptx. Replace the bracketed text. The speaker notes on each slide are your video script.
| Slide | What goes on it |
|---|---|
| 1. The problem | Team name, track and members' names. Your track's challenge in one or two plain sentences, who it matters to, and what you built |
| 2. How we used qBraid | One row per qBraid feature: what it did for your challenge, and the evidence (devices and job IDs, GPU, AI, artifact link). Delete the rows you didn't use |
| 3. What we built and found | One big visual (a chart, your tool, or a screenshot of your qBraid artifact), the result in plain words, what surprised you, what's next, and your artifact link |
Record your screen with someone on the team talking, 3 to 5 minutes. One take is fine, and it doesn't need editing. Be thorough: this video is how judges see your work, so show it off, tied to your challenge.
| Time | Show | Cover |
|---|---|---|
| 0:00 to 0:45 | Slide 1 | The problem. Your track's challenge in plain words, who it matters to (a grid operator, a utility, a security team) and what you built to tackle it. |
| 0:45 to 3:15 | Slide 2, then qBraid doing real work, live: your notebook or tool in qBraid Lab, a job on the Jobs page, a GPU run, AI Chat or the agent building something, your artifact | How you used qBraid for this problem, specifically. For each row: what you did, why it helped your challenge, and the proof. For example: "We sent our 6-substation circuit to IonQ Forte to see how hardware noise changes the placement. Here's the job, and here's how the answer shifted." |
| 3:15 to 4:30 | Your visual or tool running live, then slide 3 | What you found. The result in plain words, what you compared it with, and what surprised you. |
| Last 30 seconds | Slide 3 | What's next. What you'd try with more time. |
One rule: we have to see it run on qBraid.
Not enough on its own: "We submitted a job using qBraid." Tell us what that job did for your challenge and what you learned from it.
How to record and share:
- Mac: Cmd+Shift+5
- Windows: the Snipping Tool's record option
- Share it: upload it unlisted to YouTube, or to Vimeo, Loom or Google Drive, and set it to "anyone with the link can view"
Your credits are there to be used, so try things. Pick what moves your track's problem forward, and say in your video why you used it.
- Make it visual (we'd love to see this). An interactive view of your grid, an animation of QAOA converging, a live dashboard, or a game. Start from the one-liner above.
- Share it on Explore. Put your project in a GitHub repo and submit it from qBraid's Explore page, so anyone can launch it in qBraid Lab with one click.
- Real hardware, across devices. Run your track's problem on IonQ, Quantinuum or IQM, and compare it with the simulator. A second device should answer a question your project has, such as how hardware noise changes your answer. Prices and live status are in the device catalog.
- Run GPUs. Take your problem past what fits on a laptop.
- On-demand GPU instance: launch one from the On-Demand tab of your qBraid dashboard (GPUs in qBraid Lab).
- GPU simulators: send circuits to
qbraid:qbraid:sim:gpu-sv(exact, up to 35 qubits) orqbraid:qbraid:sim:gpu-tn(tensor network, up to 500 qubits) (GPU simulators).
- qBraid AI (encouraged). Chat and an agent do different jobs:
- Chat answers questions. Ask AI Chat in your qBraid account to compare devices, explain a result or decode an error.
- An agent does the work, inside qBraid Lab. Open the Lab Launcher and click the Claude or Codex icon. The agent runs in your Lab, so it can read your notebooks, write and run code, and build your visual. Use it when you want something built or changed, not just explained.
- Connect your own AI tool to qBraid's MCP server at
https://mcp.qbraid.com/mcp.
- The rest of the qBraid docs:
- the Circuit Composer
- Quantum Jobs in Lab
- the transpiler, which runs one circuit on any device
Have a reason for each run. Credits spent on your project count in your favor. Benchmarking a device just to run a benchmark doesn't.
These don't count on their own:
- signing up
- running local code in a notebook
- one "hello world" job unconnected to your project
- benchmarks that don't connect to your project
qBraid staff watch every video after the deadline, with your slides open. This is separate from track judging.
- Did it clearly run on qBraid? If not, the entry is out.
- How creatively did you use it? Scored 1 to 5. These all push the score up:
- a creative visualization (an interactive view, an animation or a game built on your results)
- several qBraid features, each used for a good reason
- runs across devices, or a GPU run, that move your problem forward
- a clever use of AI
We re-watch the top two and pick the winner.
Use your credits. Hardware, GPUs and qBraid AI all count when they serve your project. Benchmarks for their own sake don't.
Do we have to change our project? No. Show how qBraid fits into what you're already building.
Should we save our credits? Do your track's hardware run first. After that, spend them on your project: hardware, GPUs and qBraid AI. Stop GPU instances when you're done.
We only used simulators. Can we still win? Yes. Hardware isn't required.
Does everyone on the team need to enter? No. One person enters for the team and lists every member's qBraid email in the form, so the prize and credits reach everyone.
What's a qBraid artifact? A page you publish from qBraid Lab with qbraid artifacts publish, such as your visual.html. It gives judges a link they can open without running your code. See Make it visual.
Can we use AI? Yes, please. Say what it did and where you stepped in.
How do I know my job ran on qBraid? It's on the Jobs page in your qBraid account, with its job ID.
Questions? Ask anyone from qBraid at the event or in Discord.