Stop Paying for Claude Code — Use This FREE Tool Instead

You are helping me turn one specific YouTube video into real, durable learning.

Source video:
- Title: Stop Paying for Claude Code — Use This FREE Tool Instead
- URL: https://www.youtube.com/watch?v=9A28wWlZa8Y
- Topic: Interfaces + Open Design
- My current learning frame: Wire ZCode to a free NVIDIA-hosted model, set up a couple of backup accounts to dodge the 40-RPM limit, and build one app from a text prompt plus one from a screenshot to test both the coding and multimodal workflows.
- Why this matters: New playlist item from AI Unlocked; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:00 / Evidence 1: "What if I told you that you don't need Codex, Claude Code, or Co-work to build your own AI agent, and you could still plug in some of the strongest open-source Chinese models on the planet for free?"
- 4:32 / Evidence 2: "sensitive or private data through a free API. Second, there are rate limits, and understanding those limits is key to using these models efficiently. So, stick with me. I'll break that down in a minute. Set a password,..."
- 8:01 / Evidence 3: "click test model to confirm the connection. And there it is, connected. GLM 5.2 is officially live inside Zcode. >> Here's the number you need to know. Nvidia's free tier gives you 40 RPM, 40 requests per minute."
- 8:56 / Evidence 4: "between models freely without constantly hitting the ceiling. That one trick alone makes a huge difference in how usable this setup is. Let's actually put this to work. And this time, we're not just organizing files, we're building..."
- 11:22 / Evidence 5: "Once it's done, I'll open the file directly in the browser. And there it is, a fully working habit tracker built from a single prompt, completely free with a model that costs nothing to run. >>..."
- 14:49 / Evidence 6: "available. DeepSeek Pro was extremely slow. I mean, genuinely slow. Nematron Ultra 3 performed well, and DeepSeek V4 Flash was solid and quick. And that's it. That's exactly how you run Claude code and Codex-level agent workflows completely..."

Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable claims from the video. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A UI critique sheet for judging whether an AI interface improves control.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Intent -> Canvas -> Artifact -> Preview -> Feedback -> Iteration
   - 3 concrete examples that apply the video idea to real agentic work
   - 2 failure modes the video helps prevent
   - a checklist I can use the next time I run Codex or Claude
   - one practical exercise with a clear done signal
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.

Quality bar:
- Make this specific to "Stop Paying for Claude Code — Use This FREE Tool Instead", not a generic Interfaces + Open Design essay.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- If evidence is weak, say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.