Skip to content

docs: add README note for offline loading guidance #124

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
maod65104-hash wants to merge 1 commit into
base: main
Choose a base branch
from
Open

docs: add README note for offline loading guidance #124

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,14 @@ We recommend using this library when working with models that use the [harmony r

## Using Harmony

### Offline or network-restricted usage

If you are deploying in an offline environment or other network-restricted setup, make sure the required `.tiktoken` vocab file is available locally before calling `load_harmony_encoding` or constructing an encoding in Rust. When `TIKTOKEN_ENCODINGS_BASE` is not set, Harmony may download and cache tokenizer assets automatically; that works for connected environments, but it should not be the default expectation for air-gapped deployments.

For offline loading, place the needed vocab file (for example `o200k_base.tiktoken` or `cl100k_base.tiktoken`) in a local directory and set `TIKTOKEN_ENCODINGS_BASE` to that directory so the library can load from disk without requiring network access at runtime.

This README note only clarifies existing behavior; it does not change how vocab resolution works today.

### Python

[Check out the full documentation](./docs/python.md)
Expand Down