Amos 1 large
mukaddamzaid · Apache-2.0
English typed decisions with image inputs; distinguish zero-shot results from train-split prior fitting, and public-only Decision Index results from the full evaluation.
Browse independent decision models, publisher cards, declared licenses, inference approaches and evidence limits, including Laya, Kev, Decider, Amos and KnowLine.
15 selected models · Last curated · Download model directory JSON
These releases come from independent publishers, not TypeSafe’s Jev weights. An alternative is a candidate for comparison, not a guarantee of equivalent behavior, accuracy or interface support. Check the official documentation separately for hosted Jev versions and capabilities. TypeSafe model documentation
Each entry keeps its publisher, model-card link and a short scope note. Do not infer a release’s license from its base model: code, adapters, merged weights and data can have different terms. Availability for download and permission for commercial use are different questions; consult the publisher’s materials. Models and licenses guide
This directory does not rank releases by scores that we have not evaluated under common conditions. Match candidate sets, modalities, label sources, reasoning budgets, hardware and evaluation splits. Keep public tests, publisher-run evaluations and private-inclusive boards separate. The notes describe inspected documentation, not model execution or reproduced results. Comparison conditions
mukaddamzaid · Apache-2.0
English typed decisions with image inputs; distinguish zero-shot results from train-split prior fitting, and public-only Decision Index results from the full evaluation.
PelaAI · Apache-2.0 weights; MIT server
English/Chinese MoE decision checkpoint; public-suite checkpoint selection, disclosed task/source overlap and self-run calibration ranges constrain comparisons with independent held-out evaluations.
LycheeAI · CC BY-NC 4.0
Noncommercial Qwen-derived typed-decision model with publisher-reported public-suite results; requests exceeding context or GPU-memory limits are documented as rejected rather than truncated.
Respan · Apache-2.0
Multimodal decision model with a dedicated Blackwell Rust/CUDA runtime and optional reasoning; distinguish default serving from deterministic index settings and output-specific confidence semantics.
shlv · Qwen2.5-Omni-3B derivative · Qwen Research
One released AudioJev training seed with a Qwen Research license; original-order scores disclose audio-source overlap and are distinct from the paper’s three-seed averages.
ufak AI · Apache-2.0
Turkish typed-decision checkpoint with shipped temperature scaling. Its card flags test-informed training and worsening support-task calibration; benchmark results are not blind.
Jeva authors · license not declared
Author checkpoint with backbone, adapter and decision-head files. The inspected card declares no model license; weights have not been loaded or tested by this project.
convaiinnovations · Apache-2.0
An encoder-based decision family with English, multilingual, and workflow-specific checkpoints.
Jared Palmer · Apache-2.0
A Qwen3.5-4B-Base LoRA adapter and pointer head for decisions over documents and typed questions.
Bespoke Labs · CC BY-NC 4.0
A Qwen3.5-9B LoRA adapter returning candidate probabilities. Released under a noncommercial license.
InternLM · Apache-2.0
A decision-tuned Qwen checkpoint with a structured inference wrapper and image support.
Perplexity · Apache-2.0
A separate decision readout and saved calibration temperature. Follow the card’s attention-mode and access notes.
Fastino · Apache-2.0
A 340M English classifier with runtime label sets and single-label or multi-label decisions.
vLLM Semantic Router · Apache-2.0
A text, image, and video decision model with Choice, Yes/No, and Score outputs.
Cloudflare · Apache-2.0
A multimodal model with a joint schema head, returning probabilities over allowed answers.