Contrastive Language Models score a state against a set of candidate actions with a contrastive objective. Two projection heads on frozen Qwen3-8B embeddings, trained with InfoNCE; clm-serve maps Choice, Score and Noul onto candidate ranking.
Laya fine-tune that scans untrusted text (extracted files, knowledge-base documents, agent skills, tool descriptions) for prompt injection and data exfiltration in English and German, as a noul question answered in one forward pass. 322M parameters, PyTorch and ONNX.
Decides
choice, score, noul, rank, classify, route
noul, classify
Architecture
clm
laya
Fine-tuned from
qwen/qwen3-8b
convaiinnovations/laya-multilingual
License
apache-2.0
Unspecified; training data includes CC BY-NC-SA 4.0 material (non-commercial), not relicensed
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
95.4 ms
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
Questions
What is the difference between clm and laya-cybersec?
clm is from Contrastive-LM and laya-cybersec from TextCortex. Both have open weights you can download and run. Both answer noul and classify questions. Only clm answers choice, score, rank and route. laya-cybersec is the smaller model, at 322M parameters to 8.0B. clm is licensed apache-2.0; laya-cybersec, other.
Which is more accurate, clm or laya-cybersec?
Neither publishes an accuracy figure. Test both on your own labelled examples.
Which is cheaper, clm or laya-cybersec?
clm: Free (open weights). laya-cybersec: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run clm or laya-cybersec locally?
Yes, both: systemone pull contrastive-lm/clm and systemone pull textcortex/laya-cybersec download the weights.
p95 latency
—
149 ms
Evaluation suite
—
TextCortex saved run: 279 single-window inputs, PyTorch batch one
