Skip to content

Connect JevStyle discovery to model selection and evaluation guides #5

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

Merged
lawrence3699 merged 1 commit into from
Oct 28, 2026
Merged

Connect JevStyle discovery to model selection and evaluation guides #5

Show file tree
Hide file tree
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
13 changes: 12 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@

<!-- mcp-name: io.github.lawrence3699/jev-style -->

Local classification and routing for AI agents, with a probability for every answer. Run small models on your own machine to route support tickets, decide when to call a larger model, and review receipts or screenshots.
JevStyle is an open-source toolkit for running small decision models locally. Classify text, route requests, and score defined options with probabilities for each answer. Use it to route support tickets, decide when to call a larger model, and review receipts or screenshots with the image-capable build.

<p>
<a href="https://pypi.org/project/jev-style/"><img alt="PyPI" src="https://img.shields.io/pypi/v/jev-style?style=for-the-badge&labelColor=000000&color=0a0a0a" height="28"></a>
Expand All @@ -17,6 +17,17 @@ Local classification and routing for AI agents, with a probability for every ans

Give the model text or JSON, optional images, and typed questions. It returns yes/no probabilities, a distribution over choices, or a rating. Your application decides when to accept the result and when to ask a person or another model.

### Guides and evaluation evidence

| Start here | What it covers |
|---|---|
| [Compare local models](https://jevstyle.com/models/compare/?utm_source=github&utm_medium=readme&utm_campaign=guides) | Choose a model and backend for your inputs and hardware. |
| [Published benchmarks](https://jevstyle.com/benchmarks/?utm_source=github&utm_medium=readme&utm_campaign=guides) | Results with their evaluation scope, protocols, and limitations. |
| [Local classification and routing](https://jevstyle.com/use-cases/local-llm-routing/?utm_source=github&utm_medium=readme&utm_campaign=guides) | Score defined routes and keep a fallback for uncertain decisions. |
| [Support ticket classification](https://jevstyle.com/use-cases/support-ticket-classification/?utm_source=github&utm_medium=readme&utm_campaign=guides) | Classify a team, refund request, and urgency in one local call. |
| [Claude Code tool-call guard](https://jevstyle.com/guides/claude-code-guard/?utm_source=github&utm_medium=readme&utm_campaign=guides) | Set up a local review signal alongside existing execution controls. |
| [About JevStyle](https://jevstyle.com/about/?utm_source=github&utm_medium=readme&utm_campaign=guides) | Project identity, maintenance, licensing, and independent status. |

![The Playground answering a support-ticket request, then the agent-approval demo allowing, asking about and denying tool calls](https://raw.githubusercontent.com/lawrence3699/jev-style/main/docs/demo.gif)

## Quick Start
Expand Down
2 changes: 1 addition & 1 deletion pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
[project]
name = "jev-style"
dynamic = ["version"]
description = "Local classification and routing for AI agents, with calibrated probabilities, a systemone-compatible API, MCP tools and a Claude Code guard."
description = "Local classification, routing, and scoring for AI agents, with option probabilities, a systemone-compatible API, MCP tools and a Claude Code guard."
readme = "README.md"
license = "Apache-2.0"
requires-python = ">=3.10"
Expand Down
Loading