中文Sign up

FrontiersMind/Lumma-fev-9b

🤗 Hugging Face sourcetext-classificationapache-2.07.9B params16 GBsafetensors✓ 2 checksumsupdated today
Submit in one command

Run it next to your model folder. It makes the torrent, checks your files against Hugging Face, and submits it. You just start seeding and paste your key from your account. It only reads your files and never changes them. Read the script first if you like.

curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo FrontiersMind/Lumma-fev-9b ./model-folder
Needs a seeder →

Lumma-Fev-9B

Lumma-Fev-9B is a decision model. It reads one document (the state) and a set of typed questions, and returns a probability distribution for each question in a single forward pass. It never generates text, so there is nothing to parse and nothing to hallucinate.

Benchmarks

Benchmark / Metric TypeSafe Jev 1.13.0 Laya Gliner2.5-Decide Lumma-Fev-0.15B Lumma-Fev-0.6B Lumma-Fev-4B Lumma-Fev-9B
Typed-decisions 0.72 0.76 0.46 0.49 0.64 0.78 0.81
AG News 0.91 0.95 0.48 0.89 0.84 0.93 0.95
DAIR Emotion 0.48 0.595 0.55 0.68 0.89 0.94 0.94
Banking77 0.87 0.425 0.68 0.47 0.90 0.93 0.94
Average 0.75 0.68 0.54 0.63 0.82 0.90 0.91
P50 latency 256 ms 32.8 ms 99.85 ms 35.576 ms 45.82 ms 175.93 ms 269.39 ms
Weights Close Open Open Open Open Open Open
Cost $0.042 / 1M tokens $0 self-hosted $0 self-hosted $0 self-hosted $0 self-hosted $0 self-hosted $0 self-hosted

Use it with transformers

from transformers import AutoModel
import torch

model = AutoModel.from_pretrained("FrontiersMind/Lumma-fev-9b", trust_remote_code=True,dtype=torch.bfloat16)   # the tokenizer is loaded with it

answers = model.decide(
    state="My running shoes arrived in the wrong size. Can I exchange them for a size 10?",
    questions={
    "department": {
        "type": "choice",
        "instructions": "Which team should handle this?",
        "criteria": {
            "returns": "Exchanges, wrong items, damaged items",
            "shipping": "Delivery status, delays, lost packages",
            "billing": "Charges, invoices, payment problems",
        },
    }
}
)
print(answers)

state can be text, a JSON object or an array. Pin a version with revision="<commit>", and move the model to a GPU with model.to("cuda") (bf16 is the stored precision).

Use it with the lumma-fev package

pip install lumma-fev              # local inference
pip install "lumma-fev[serve]"     # plus the API server

import lumma_fev

model = lumma_fev.load("FrontiersMind/Lumma-fev-9b")        # picks cuda, mps or cpu
print(model.decide("Two charges on my card for one order.",
                   {"billing": {"type": "noul", "instructions": "Is this about billing?"}}))

Serve it as an API

lumma-fev-serve exposes the TypeSafe POST /v1/systemone contract, so existing TypeSafe clients work by changing their base URL.

lumma-fev-serve --model FrontiersMind/Lumma-fev-9b --host 0.0.0.0 --port 8000
# optional: LUMMA_FEV_API_KEY=<key> requires "Authorization: Bearer <key>"; --cors for browser apps

curl -s localhost:8000/v1/systemone -H 'content-type: application/json' -d '{
  "state": "I was charged twice. Please fix this ASAP.",
  "questions": {"billing": {"type": "noul", "instructions": "Is this ticket about billing?"}}
}'

From Python, with the lumma-fev client or the TypeSafe SDK:

from lumma_fev import Client

client = Client("http://127.0.0.1:8000")
print(client.decide("I was charged twice.", {"billing": {"type": "noul", "instructions": "Is this about billing?"}}))

from typesafe_sdk import Noul, TypeSafeClient     # pip install typesafe-sdk

with TypeSafeClient(api_key="local", base_url="http://127.0.0.1:8000") as ts:
    print(ts.system_one(state="I was charged twice.", questions={"billing": Noul(instructions="Is this about billing?")}).nouls["billing"].noul)

Answers

Type Criteria Answer
noul optional {"true": ..., "false": ...} noul: probability of yes
choice {name: description or null} choice (most likely name), confidence, probabilities by name
score ordered list of levels score (expected level), confidence, legend, probabilities by level

Questions never see each other: each one reads the state and its own instructions and options only, so one question cannot change another's answer. Text inside the request cannot forge the model's delimiter tokens.

Limits

  • confidence and probabilities are the model's own estimates. Measure calibration on your own labelled data before gating automated actions on them.

NOTE: This Model is continual pre-trained and fine-tuned on top Qwen-3.5.-4B Series.

Feedback