jeb:27b
4 TagsUpdated 27B params4096 contextEnglishApache-2.0by frontier-infra
Jebadiah 27B (Qwen3.8), Q4_K_M GGUF, 17 GB: the largest Jeb, for a 24 GB GPU.
27b
ollaya run jeb:27b --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": "jeb:27b",
"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": "jeb:27b",
"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: "jeb:27b",
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
- weights91b4c7ab57ce · 16.8 GB
gguf · Q4_K - Medium · huggingface.co/frontier-infra/jebadiah-27b-GGUF/resolve/7451e61…/jebadiah-27b-Q4_K_M.gguf - decision055dcab50720 · 7 KB
{"engine": "llama", "family": "jebadiah", "layout": "jebadiah-v1", "gguf": {…}, …} - calibration4940c922a807 · 162 B
{"temperature": [1.232, 0.756, 1.297]} - licensebf69a57a4975 · 10 KB
Jebadiah by Jason Brashear and AINode (https://huggingface.co/frontier-infra, https://github.com/getainode/jebadiah): rank-16 LoRA merges into Qwen3.5-4B, Qwen3.5-9B and Qwen3.8-27B by the Qwen team (Apache-2.0), published by the authors as GGUF.
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.