# ZeroPoint decision model by Javier Marín

> ZeroPoint scores options against an implicit none state on a CPU. Read its bundled model, short input budgets and saved missing-option tests.

Updated 2026-11-12 (Australia/Adelaide)

## Quick answers

### What is ZeroPoint by Javier Marín?

[ZeroPoint](<https://decisionmodels.ai/models/zeropoint>): A CPU decision model that scores supplied options against a fixed zero-energy none state.

[ZeroPoint facts](<https://decisionmodels.ai/models/zeropoint>)

[Source](<https://github.com/Javihaus/zeropoint>)

### Who makes ZeroPoint?

[ZeroPoint](<https://decisionmodels.ai/models/zeropoint>): Javier Marín

[ZeroPoint facts](<https://decisionmodels.ai/models/zeropoint>)

[Source](<https://github.com/Javihaus/zeropoint>)

By Javier Marín · First published package release; October catch-up: 2026-10-31 · OPEN

A CPU decision model that scores supplied options against a fixed zero-energy none state. It packages an encoder derived from multilingual-e5-small, ONNX weights and a fitted head. Its runtime sorts options into a canonical order before scoring, so order consistency belongs to the complete serving path.

## Facts

- Size: 89 MB model files, per the maker
- Inputs: Text and option labels; no free-form generation
- Limits: Reads 192 text word pieces and 16 per option, with truncation flags; options processed in blocks of 20
- Access: Bundled model files; Python with NumPy, ONNX Runtime and tokenizers
- Licence: Apache-2.0 code; upstream multilingual-e5-small is MIT, with notices

## Claims and sources

- **VENDOR-REPORTED**: The maker reports 53.6% correct decisions including none across 900 matched cases, half with no right option, versus 58.6% for [Strands Decider 2B v1](<https://decisionmodels.ai/models/strands-decider>). [ZeroPoint](<https://decisionmodels.ai/models/zeropoint>) takes a 42 ms median on a two-core CPU with two threads; Strands uses an NVIDIA L4 in that comparison. These timings do not compare raw speed on the same hardware. [source](<https://github.com/Javihaus/zeropoint/blob/main/benchmarks/compare/results/tables.md>)

## Links

- [Source: Code and bundled model](<https://github.com/Javihaus/zeropoint>)
- [Source: Saved comparison](<https://github.com/Javihaus/zeropoint/blob/main/benchmarks/compare/results/tables.md>)
- [Source: Licence](<https://github.com/Javihaus/zeropoint/blob/main/LICENSE>)
- [Source: Model manifest](<https://github.com/Javihaus/zeropoint/blob/main/src/zeropoint/model/manifest.json>)
- [Source: Package releases](<https://github.com/Javihaus/zeropoint/releases>)

## Related projects

### ZeroPoint CPU decisions

[ZeroPoint CPU decisions](<https://decisionmodels.ai/projects#project-zeropoint>)

[Source: Javihaus/zeropoint](<https://github.com/Javihaus/zeropoint>)

Scores option labels and an implicit none state using a bundled ONNX encoder and fitted head. The maker publishes saved comparisons, model hashes and fitting code. Canonical option sorting and one-case inference are part of its consistency recipe. The 900-case matched comparison reports 53.6% correct including none, with half the cases having no right option. Short input budgets and truncation flags remain explicit.

- Creator: Javier Marín

- Stars: 12

- Licence: Apache-2.0

- Model: [ZeroPoint](<https://decisionmodels.ai/models/zeropoint>)

- Tags: decision-model

- [Source: Licence](<https://github.com/Javihaus/zeropoint/blob/main/LICENSE>)
- [Source: Saved run tables](<https://github.com/Javihaus/zeropoint/blob/main/benchmarks/compare/results/tables.md>)