CalibratedDecisions.

Repo · Routing & model choice

JEV-MLX

MLX local action selection from existing LLMs, with no extra training or decision-specific weights.

Open github.com ↗

From the repository

Typed decisions from existing MLX-LM and MLX-VLM models on Apple Silicon. Text + images, no fine-tuning or extra decision weights. Related approaches: AnyJev, OpenDecider, Laya.

action-selectionapple-silicondecision-modeljevllmlocal-inferencelocal-llmmlxmlx-lmmlx-vlmmultimodalnatural-language-processingon-device-aipythonsemantic-routingtraining-freetyped-decisionsvision-language-model

How builders describe it

MLX local action selection from existing LLMs, with no extra training or decision-specific weights. Semantic routing on Apple Silicon; compare OpenDecider and Laya.

The decision Jev makes

Pick an agent, a model, a skill from a catalog, or a function and its arguments.

Where it fits

Deciding which agent, model, skill or function should handle a request. Jev is a natural router: the options are fixed and the confidence tells you when to fall back. All 348 routing & model choice projects →

Related projects