laya:typed-decisions-fp32
10 TagsUpdated 421M params1024 contextEnglishApache-2.0by Convai Innovations
Fine-tuned on typed-decisions workflows: 0.766 accuracy vs 0.727 published for Jev 1.13.
fine-tuned421m
ollaya run laya:typed-decisions-fp32 --preset triage "I was charged twice for my subscription this month and want a refund."curl http://localhost:11435/api/decide \
-H "Content-Type: application/json" \
-d '{
"model": "laya:typed-decisions-fp32",
"state": "I was charged twice for my subscription this month and want a refund.",
"questions": {
"department": {
"type": "choice",
"instructions": "Which team should handle this?",
"criteria": {
"billing": "Payments, invoices and refunds",
"technical": "Bugs, errors and outages",
"account": "Login, profile and settings"
}
},
"refund": {
"type": "noul",
"instructions": "Is the customer asking for a refund?"
}
}
}'# Already using a TypeSafe SDK? Set TYPESAFE_BASE_URL=http://localhost:11435 instead.
import requests
response = requests.post(
"http://localhost:11435/api/decide",
json={
"model": "laya:typed-decisions-fp32",
"state": "I was charged twice for my subscription this month and want a refund.",
"questions": {
"department": {
"type": "choice",
"instructions": "Which team should handle this?",
"criteria": {
"billing": "Payments, invoices and refunds",
"technical": "Bugs, errors and outages",
"account": "Login, profile and settings"
}
},
"refund": {
"type": "noul",
"instructions": "Is the customer asking for a refund?"
}
}
},
)
answers = response.json()["answers"]
print(answers["department"]["choice"], answers["refund"]["noul"])const response = await fetch("http://localhost:11435/api/decide", {
method: "POST",
headers: { "Content-Type": "application/json" },
body: JSON.stringify({
model: "laya:typed-decisions-fp32",
state: "I was charged twice for my subscription this month and want a refund.",
questions: {
department: {
type: "choice",
instructions: "Which team should handle this?",
criteria: {
billing: "Payments, invoices and refunds",
technical: "Bugs, errors and outages",
account: "Login, profile and settings"
}
},
refund: {
type: "noul",
instructions: "Is the customer asking for a refund?"
}
}
}),
});
const { answers } = await response.json();
console.log(answers.department.choice, answers.refund.noul);Details
- graph7d648ef513e8 · 4 MB
onnx · ModernBERT-large · 421M · fp32 - grapha4e9008a387a · 4 MB
onnx · ModernBERT-large · 421M · fp16 - weights4fa56de72383 · 843 MB
huggingface.co/convaiinnovations/laya/resolve/aa8c91c…/typed-decisions/model.safetensors - tokenizer6c8aaa9a5420 · 4 MB
huggingface.co/convaiinnovations/laya/resolve/aa8c91c…/typed-decisions/tokenizer/tokenizer.json - decisiona1d3fbc58178 · 479 B
{"engine": "onnx", "family": "laya", "encoder": "answerdotai/ModernBERT-large", "layout": "laya-markers-v1", …} - calibrationcac4b5f486ad · 353 B
{"temperature_by_options": {"choice:3-5": …, "choice:6-10": …, "score:3-5": …, "noul:2": …, "choice:11+": …, "choice:2": …}} - licenseec56c3e69fab · 10 KB
Laya by Convai Innovations (https://huggingface.co/convaiinnovations/laya) - paramsdd15c90375b8 · 21 B
precision fp32
Every layer is checked against its sha256 when it is pulled. Weights and tokenizers download from the model author's Hugging Face repository at a pinned commit; Ollaya never re-hosts them.