winnex-nano
Native Winnex inference engine — deterministic spectral tokenizer, multimodel weight balancing, and chunk streaming via the Madhava fast build.
Model-agnostic (Qwen / BERT / DeepSeek / GPT), C++20/OpenCL, no CUDA. Reuses the winnex-madhava engine kernels (QKᵀ matmul, selective-top-K with Cauchy-Schwarz bounds) — no code duplication.
Components
| Module | Purpose |
|---|---|
SpectralTokenizer |
Character → quaternion spectrum (deterministic, no BPE, no vocabulary) |
WeightBalancer |
Multimodel fusion: W' = Σᵢ αᵢ·R(qᵢ)·Wᵢ |
StreamEngine |
Chunk generation via the Madhava fast build (working memory) |
server_sse |
OpenAI-compatible /v1/chat/completions (stream + non-stream) |
Installation
pip install winnex-nano
Requires winnex-madhava>=1.8.1 (the engine), numpy. Optional: winnex-nano[server]
for the SSE server (fastapi, uvicorn, httpx).
Quick start
import winnex_nano as wn
# 1. Deterministic spectral tokenizer (no BPE, no vocabulary)
tok = wn.SpectralTokenizer(embed_dim=64)
states = tok.encode("Winnex AI") # list of quaternions
text = tok.decode(states) # round-trip: "Winnex AI"
# 2. Multimodel weight balancing: W' = Σ αᵢ·R(qᵢ)·Wᵢ
bal = wn.WeightBalancer()
model_a = {"layer.0.weight": [1, 2, 3, 4]}
model_b = {"layer.0.weight": [10, 20, 30, 40]}
a = wn.BlendWeight(); a.alpha = 0.5
b = wn.BlendWeight(); b.alpha = 0.5
fused = bal.blend([model_a, model_b], [a, b]) # [5.5, 11, 16.5, 22]
# 3. Streaming via the Madhava fast build
engine = wn.StreamEngine()
chunks = []
engine.stream(
"O Madhava e deterministico.",
lambda ctx, top: " chunk" if len(chunks) < 3 else "",
lambda c: chunks.append(c.text) if not c.done else None,
)
SSE server (OpenAI-compatible)
pip install "winnex-nano[server]"
python -m winnex_nano.server_sse 30002
curl -N http://localhost:30002/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"messages":[{"role":"user","content":"O Madhava e deterministico"}],"stream":true}'
Honest benchmark (vs BPE)
Run the full benchmark on Kaggle — installs winnex-nano from PyPI and
reports round-trip, throughput and O(1) decode scaling on real multilingual
text:
| Metric | winnex-nano spectral | BPE (Qwen2.5) |
|---|---|---|
| Round-trip perfect | 8/8 | n/a |
| Char error rate | 0.00% | n/a |
| Encode throughput | 234K chars/s | 4.3M chars/s |
| Stream (3 chunks) | 1.7 ms | ~2.1 s (HTTP) |
| Vocabulary | none | 151K tokens |
The spectral tokenizer is 100% deterministic with no vocabulary dependency, but 50–70× slower and ~1000× less compact than BPE. It is an autonomous encoding for the native multimodel engine, NOT a drop-in BPE replacement for a BPE-trained model.
License
Business Source License 1.1 (BSL 1.1) — pay@winnex.ai.
Pre-patent: https://zenodo.org/records/21861809
1 maintainer
Metadata
1 maintainer
Release files for winnex-nano 0.1.13
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 | Reset |
|---|---|---|---|---|
| winnex_nano-0.1.13-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl | CPython 3.12 | CPython 3.12 | Linux glibc 2.28+ x86-64, Linux glibc 2.27+ x86-64 | Details |
Release files / winnex_nano-0.1.13-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
| Download URL | winnex_nano-0.1.13-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 450.5 kB |
| Tags | CPython 3.12 Linux glibc 2.27+ x86-64 Linux glibc 2.28+ x86-64 |
|
SHA-256 checksum How to use checksums |
a41c581e6d5dab3e5dcfea1798d572fb67ffae351ca86489372fed171544d421
|
|
BLAKE2b-256 checksum How to use checksums |
d18bbc331f5248436b368971989b859f8e28eb2c500473cb7324665571a3e470
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
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.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 19, 2026.
Transparency logSecurity for winnex-nano 0.1.13
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.
Full provenance
All 1 file in this release has been uploaded with provenance metadata.
Attestation for winnex_nano-0.1.13-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Oct 19, 2026.
Transparency logA 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.
