Skip to content

docs: Rust model parsing guide — parse .fbz models without Python - #27

Open
ecoPrimal wants to merge 1 commit into
Brainchip-Inc:mainfrom
ecoPrimal:rust-parsing
Open

ecoPrimal wants to merge 1 commit into
Brainchip-Inc:mainfrom
ecoPrimal:rust-parsing

Conversation

@ecoPrimal

Copy link
Copy Markdown

Summary

Adds a rust/ directory with documentation on how to parse all .fbz models from this zoo using pure Rust via syntheticChemistry/rustChip.

No Python, no SDK, no MetaTF required. The parser handles Snappy decompression and FlatBuffer parsing natively.

Why

Currently, loading models requires the Python SDK (akida package). The Rust parser provides an alternative for:

  • Embedded systems where Python is not available
  • CI/CD pipelines that need fast model validation
  • Production deployments where a single static binary is preferred
  • Systems running Linux > 6.9 where the C kernel driver does not build

Validated

All v1 and v2 models from this zoo parse correctly (tested April 2026, re-validated October 2026).

Related

Made with Cursor

Shows how to parse all .fbz models from this zoo using rustChip
(syntheticChemistry/rustChip). Pure Rust, no Python, no SDK.

Validated against all v1 and v2 models.

Co-authored-by: Cursor <cursoragent@cursor.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant