Q-Star AI website
Public English platform homepage and application narrative at ai.q-star.ink.
Evidence pack
Proof that Q-Star AI is already operating AI generation workflows.
Q-Star AI is building a multi-agent AI creation infrastructure for video, apps, and automated creative workflows.
This page organizes the application evidence Google should see: deployed systems, infrastructure traces, generated outputs, workload metrics, and the migration path to Google Cloud.
Active
AI workflows across video, app generation, agents, and media production
Live
Public web endpoints and hosted generated media assets
Ready
Evidence checklist for logs, screenshots, demos, and outputs
Next
Gemini API, Cloud Storage, and Cloud Run migration activity
Existing deployed systems.
The application should show that Q-Star AI already has real systems and workflows. Screenshots do not need to be polished; real operating evidence is more valuable than marketing visuals.
OpenClaw / AgentOS
Agent planning, task execution, model routing, retry, and workflow review screenshots.
QClip / media generation
Prompt input, generation jobs, rendering logs, output media, and storage evidence.
Drama Engine
Short-drama story planning, scenes, characters, generated scripts, and final clips.
APK Generator
App generation prompts, build logs, APK artifacts, and generated mobile UI examples.
Hermes / model workflows
Multi-model routing traces, provider calls, token usage, and agent orchestration logs.
Captured evidence snapshot.
This is the first verified infrastructure snapshot captured from the current 202 server on May 7, 2026. It proves there is already a deployed public system with live files, sample outputs, and running backend services.
| Evidence item | Verified state | Application value |
|---|---|---|
| Public site | https://ai.q-star.ink/ returns HTTP 200 and serves the Q-Star AI landing page. | Shows a deployed English platform presence for reviewers. |
| Evidence page | https://ai.q-star.ink/evidence.html and /evidence both return HTTP 200. | Creates a direct reviewer path from homepage to evidence material. |
| Nginx deployment | Server name ai.q-star.ink / stu.q-star.ink, root /var/www/ai-qstar, HTTPS enabled by Certbot. | Shows current production-style web infrastructure. |
| Running service | Nginx is active and enabled; configuration test passes. | Supports reliability and operations claims. |
| Deployed modules | Existing pages include agent.html, creative.html, drama-service.html, ecom-video.html, insight.html, and shop.html. | Shows Q-Star AI has multiple deployed product surfaces. |
| Generated media outputs | Sample videos are deployed under /var/www/ai-qstar/videos, including Douyin, Pinduoduo, Taobao, and Xiaohongshu samples. | Supports the AI video and generated output narrative. |
| Docker workloads | Running containers include huobao-drama, qstar-cobalt, yt-session-generator, and kokoro-tts. | Shows active backend and media-related workload infrastructure. |
| Server capacity | Server has 329 GB disk with 173 GB available and 15 GiB memory. | Provides baseline capacity evidence before Google Cloud migration. |
Infrastructure evidence to capture.
Google Cloud reviewers should see that the current workloads require real compute, storage, queues, logging, monitoring, and model usage.
- Nginx configuration and public deployment path for ai.q-star.ink.
- Docker containers, systemd services, or process manager screens.
- Agent workflow logs showing prompt decomposition and task execution.
- Queue activity for async dispatch, retries, rendering, and worker progress.
- GPU jobs, render workers, video/image/audio processing logs.
- Model API logs showing text, code, planning, and multimodal generation calls.
- Storage screenshots showing generated videos, images, APKs, and prompt archives.
- Monitoring views for request volume, latency, errors, health checks, and cost tracking.
Generated outputs.
The strongest evidence is a prompt-to-output chain: prompt, agent plan, model routing, generation logs, rendered assets, and final output URL.
| Output type | Evidence to attach | Why it matters |
|---|---|---|
| AI video | Prompt, render job, output video URL, storage path | Shows media-heavy workloads and GPU/rendering demand |
| Short drama | Story plan, scene outputs, generated trailer, asset folders | Shows multimodal generation and production pipelines |
| AI app generation | Prompt, generated UI/code, build log, APK artifact | Shows code generation, packaging, and repeatable app workflows |
| Agent planning | Task graph, model selection, worker logs, completion state | Shows the platform is a multi-agent workflow system |
| Workflow output | Audit trail, retries, cost estimate, generated files | Shows reliability, observability, and infrastructure need |
Current workload and projection evidence.
The numbers should be backed by logs or screenshots. Current projections are based on internal workloads, active demos, and a planned early-access rollout.
| Metric | Current baseline | 6-month projection | 12-month projection | Evidence source |
|---|---|---|---|---|
| Daily generations | 300 | 20,000 | 80,000 | API logs, generation database, workflow records |
| API requests / day | 5,000 | 350,000 | 1,500,000 | Gateway logs and server metrics |
| Async jobs / day | 1,000 | 120,000 | 500,000 | Queue logs and task database |
| Token consumption / month | 60M | 2B | 8B | Model provider dashboards |
| Video rendering minutes / month | 1,200 | 80,000 | 300,000 | Render worker logs |
| GPU inference hours / month | 120 | 3,000 | 10,000 | GPU server logs and job duration reports |
| Generated storage | 0.5 TB | 12 TB | 50 TB | Storage usage metrics |
Google Cloud migration plan.
The first migration step should create real Google Cloud activity before the application is submitted: Gemini API, Cloud Storage, and Cloud Run are the highest-value starting points.
1
Gemini API
Route planning, code, text, and multimodal generation tasks through Gemini where appropriate.
2
Cloud Storage
Store generated videos, images, APKs, prompt archives, workflow artifacts, and demo outputs.
3
Cloud Run
Deploy API services, agent workers, routing endpoints, and lightweight workflow services.
4
Scale layer
Add Pub/Sub, BigQuery, Monitoring, GPU workers, and GKE as workload volume grows.
Next evidence gap: screenshots and demo video.
The website and application pack are in place. The highest-value next work is collecting real system screenshots and recording a two-minute prompt-to-output demo.