Models

Published checkpoints and lineage

Each model states its architecture, training data, evaluation scope and exact publication status. Native and CoderSec models are independent and must not be treated as one architecture family.

ModelLineageParametersTokenizerStatusArtifact
XRFM Native 25MOriginal, random initialization25,200,000T4 32K BPEPublishedHugging Face
XRFM Native 85MOriginal, random initialization86,446,464T4 32K BPEPublishedHugging Face
XRFM CoderSecQwen3-1.7B base + QLoRA~1.7B baseQwen tokenizerNot publishedSee status page

Native weights

Released under Apache-2.0 metadata. Repository source remains MIT-licensed; the weight license is artifact-specific.

CoderSec attribution

Base model Qwen/Qwen3-1.7B revision 70d244cc, Apache-2.0. No adapter or merged weights are public.

No hidden lineage

Native models loaded no third-party pretrained weights. CoderSec did, and says so on every public surface.