I re-downloaded the latest version of TinyTorch. To maintain consistency with my previous progress, I directly used the commands: tito module start 01, tito module complete 01, tito module start 02, tito module complete 02, tito module start 013, tito module complete 03. However, it failed during testing—this is the same issue I encountered yesterday.
Originally created by @wz1114841863 on GitHub (Jan 14, 2026).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/1110
Originally assigned to: @profvjreddi on GitHub.
I re-downloaded the latest version of TinyTorch. To maintain consistency with my previous progress, I directly used the commands: `tito module start 01`, `tito module complete 01`, `tito module start 02`, `tito module complete 02`, `tito module start 013`, `tito module complete 03`. However, it failed during testing—this is the same issue I encountered yesterday.
<img width="709" height="424" alt="Image" src="https://github.com/user-attachments/assets/72ff9f57-6c9d-4af1-82f2-aa89f79fa926" />
<img width="1024" height="854" alt="Image" src="https://github.com/user-attachments/assets/9a5702bc-e0c4-466b-a982-3a6d2fca75f1" />
Thanks for reporting this @wz1114841863! You've found a real bug in the workflow.
Root Cause: The integration tests were running BEFORE the module was exported. Since integration tests import from tinytorch.core.layers (which only exists after export), they always failed on a fresh install.
The Fix: I've reordered the workflow in commit 1f7a2bb:
Old Order
New Order
1. All tests
1. Unit tests
2. Export
2. Export
3. Track progress
3. Integration tests
4. Track progress
Now:
Unit tests run first (they test your source files directly)
Export happens next (creates tinytorch/core/layers.py)
Integration tests run last (they can now find the exported module)
To get the fix: Re-download the latest version:
curl -fsSL https://tinytorch.ai/install | bash
I will do a new bug fix release once all the tests pass.
Then try tito module complete 03 again. Let me know if you hit any other issues!
<!-- gh-comment-id:3749908351 -->
@profvjreddi commented on GitHub (Jan 14, 2026):
Thanks for reporting this @wz1114841863! You've found a real bug in the workflow.
**Root Cause:** The integration tests were running BEFORE the module was exported. Since integration tests import from `tinytorch.core.layers` (which only exists after export), they always failed on a fresh install.
**The Fix:** I've reordered the workflow in commit [1f7a2bb](https://github.com/harvard-edge/cs249r_book/commit/1f7a2bb82):
| Old Order | New Order |
|-----------|-----------|
| 1. All tests | 1. Unit tests |
| 2. Export | 2. Export |
| 3. Track progress | 3. Integration tests |
| | 4. Track progress |
Now:
- Unit tests run first (they test your source files directly)
- Export happens next (creates `tinytorch/core/layers.py`)
- Integration tests run last (they can now find the exported module)
**To get the fix:** Re-download the latest version:
```bash
curl -fsSL https://tinytorch.ai/install | bash
```
I will do a new bug fix release once all the tests pass.
Then try `tito module complete 03` again. Let me know if you hit any other issues!
v0.1.4 is now released with this fix! See #1112 for update instructions.
<!-- gh-comment-id:3787478495 -->
@profvjreddi commented on GitHub (Jan 23, 2026):
v0.1.4 is now released with this fix! See #1112 for update instructions.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @wz1114841863 on GitHub (Jan 14, 2026).
Original GitHub issue: https://github.com/harvard-edge/cs249r_book/issues/1110
Originally assigned to: @profvjreddi on GitHub.
I re-downloaded the latest version of TinyTorch. To maintain consistency with my previous progress, I directly used the commands:

tito module start 01,tito module complete 01,tito module start 02,tito module complete 02,tito module start 013,tito module complete 03. However, it failed during testing—this is the same issue I encountered yesterday.@profvjreddi commented on GitHub (Jan 14, 2026):
Thanks for reporting this @wz1114841863! You've found a real bug in the workflow.
Root Cause: The integration tests were running BEFORE the module was exported. Since integration tests import from
tinytorch.core.layers(which only exists after export), they always failed on a fresh install.The Fix: I've reordered the workflow in commit 1f7a2bb:
Now:
tinytorch/core/layers.py)To get the fix: Re-download the latest version:
I will do a new bug fix release once all the tests pass.
Then try
tito module complete 03again. Let me know if you hit any other issues!@profvjreddi commented on GitHub (Jan 23, 2026):
v0.1.4 is now released with this fix! See #1112 for update instructions.