enigma-turbo-model
This package is the model that the experimental TurboStrategy in the Enigma Python SDK, enigma-api-client, runs on your machine to turn a question into a GraphQL query for the Enigma API without an LLM call. Install it through the SDK's turbo extra, which also installs the runtime it needs:
pip install "enigma-api-client[turbo]"
The Python SDK guide's TurboStrategy section shows how to use the strategy, and its Choose a strategy section gives the time each strategy takes to write a query.
The package installs one folder holding these data files:
model.onnx: a 32M-parameter encoder fine-tuned from Ettin (jhu-clsp/ettin-encoder-32m), with int8 weights for ONNX Runtime.tokenizer.json: Ettin's tokenizer.meta.json: the templates and fields the model gives, in its output order.city_state.json: US city names with their states and populations, from the US Census Bureau's Vintage 2024 population estimates for places of 5,000 people or more.words.txt: the web2 word list (Webster's Second International Dictionary, 1934), which splits a website's domain into words.
Each SDK release pins the model and ONNX Runtime versions its tests check. The model gives the same replies on the x86 CPUs with AVX-512 it was tested on; other CPUs, ARM and x86 without AVX-512 among them, can reply differently to some prompts.
License
The package is licensed under the Apache License 2.0 (LICENSE). model.onnx and tokenizer.json derive from Ettin, licensed under the MIT License (LICENSE-ETTIN). city_state.json derives from US Census Bureau data, a work of the US government. The web2 list's supplier states that its 1934 copyright has lapsed.
Metadata
Release files for enigma-turbo-model 0.2.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| enigma_turbo_model-0.2.2-py3-none-any.whl | Python 3 | none | any | Details for enigma_turbo_model-0.2.2-py3-none-any.whl |
Release files / enigma_turbo_model-0.2.2-py3-none-any.whl
| Download URL | enigma_turbo_model-0.2.2-py3-none-any.whl |
|---|---|
| Size | 26.6 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
16164ef8a0dc9c8799dbd9a44cf6a946619f5853fb05bd780dc972414400d821
|
|
BLAKE2b-256 checksum How to use checksums |
be8423fb67da51f09978a2d1cbf040d5a4ee3bbb0a4650fdc10d1f79abdfc7a0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.13.13
|
Security for enigma-turbo-model 0.2.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.