Qwen/Qwen2.5-1.5B-Instruct (Base Model)
βββ AnkitNayak-dev/QJev-1.5B (Flagship Fused 16-bit, 3.08 GB)
βββ AnkitNayak-dev/QJev-1.5B-LoRA (PEFT Adapter, 20 MB)
βββ AnkitNayak-dev/QJev-1.5B-4bit (NF4 Quantized, 1.15 GB)
Overview β’ Benchmark Evaluation β’ Model Family β’ Quickstart β’ Serving and Deployment β’ Architecture β’ Repository Structure β’ Citation
Traditional large language models perform decision-making through sequential, autoregressive token generation. This process incurs substantial latency (often 300 to 600 ms) and introduces non-deterministic sampling variance when parsing formatted outputs.
QJev-1.5B implements a direct System 1 decision mechanism by coupling a calibrated Multi-Layer Perceptron (MLP) decision head to the final hidden representation of Qwen/Qwen2.5-1.5B-Instruct. Rather than generating text tokens, the model computes a calibrated probability distribution across up to 10 candidate actions in a single forward pass (~31 ms on consumer hardware).
QJev-1.5B is evaluated against the official JevBench metamorphic test harness (jevbench-mini, 240 evaluation cases, 50 scored relations, 1,468 trials) following standard BenchmarkHeaven submission criteria.
All models evaluated under official strict calibration criteria (tolerance = 0.05):
| Model | Parameters | Official Score (tol=0.05) | Category | Inference Latency | Memory Footprint |
|---|---|---|---|---|---|
EXAONE-1.2B |
1.2B | 47.7% | Sub-2B | ~25 ms | ~2.8 GB |
SargeDev/jev-gate-student-b |
0.49B | 50.2% | Sub-2B (Binary Only) | ~23 ms | ~1.5 GB |
decision-2b |
2.0B | 54.8% | 2B Tier | ~40 ms | ~4.5 GB |
pjev-2b |
2.0B | 56.5% | 2B Tier | ~38 ms | ~4.4 GB |
QJev-1.5B (Ours, Phase 3) |
1.54B | 62.53% | Rank 1 Sub-2B | ~31 ms | 3.4 GB (1.07 GB in 4-bit) |
jiwo-4b |
4.0B | 60.2% | 4B Tier | ~65 ms | ~8.8 GB |
decision-4b-v12 |
4.0B | 60.4% | 4B Tier | ~62 ms | ~8.5 GB |
autotrust/JEV-9B |
9.0B | 72.0% | 9B Tier | ~85 ms | ~18.5 GB |
autotrust/JEV-27B |
27.0B | 76.5% | Datacenter Tier | ~220 ms | ~55.0 GB |
Tolerance specification: The official BenchmarkHeaven evaluation suite strictly uses tolerance = 0.05 for headline Capability scores. Under exploratory relaxed evaluation (tolerance = 0.10), QJev-1.5B achieves 71.50%.
| Dimension | Baseline (Pre-Consistency) | QJev-1.5B (Phase 3 Consistency) | Delta | Formal Property Evaluated |
|---|---|---|---|---|
BAT.isolation |
100.0% | 100.0% | 0.0% | Invariance to batch composition and sample ordering |
REP.type |
77.1% | 83.3% | +6.2% | Invariance across prompt format representations (MCQ, boolean, ordinal) |
REP.invariance |
73.2% | 74.4% | +1.2% | Stability under option key renaming and syntax perturbations |
LOG.bounds |
62.5% | 67.9% | +5.4% | Adherence to probabilistic upper and lower bounds |
CHO.conditioning |
41.7% | 56.9% | +15.2% | Preservation of relative odds upon elimination of irrelevant options |
REP.equivariance |
48.6% | 55.6% | +7.0% | Consistent probability shifts under negation transformations |
MEA.partition |
30.7% | 41.1% | +10.4% | Probability mass conservation under option clustering |
| Overall (Official Strict 0.05) | 58.67% | 62.53% | +3.86% | Standard JevBench Capability Benchmark |
Machine-readable evaluation artifacts are archived under reports/jevbench/.
QJev-1.5B is available in three deployment formats:
| Format | Hugging Face Repository | Disk Size | VRAM Footprint | Recommended Use Case |
|---|---|---|---|---|
| Full Fused (FP16) | AnkitNayak-dev/QJev-1.5B | 3.08 GB | 3.4 GB | Production servers, vLLM, standalone microservices |
| PEFT Adapter | AnkitNayak-dev/QJev-1.5B-LoRA | 20.7 MB | Shared | Multi-tenant deployments caching Qwen2.5-1.5B-Instruct |
| NF4 Quantized | AnkitNayak-dev/QJev-1.5B-4bit | 1.15 GB | 1.07 GB | Resource-constrained edge instances, local workstations |
git clone https://github.com/AnkitNayak-dev/QJev-1.5B.git
cd QJev-1.5B
pip install -r requirements.txt
from qjev import QJev
# Load any model variant directly from Hugging Face Hub
model = QJev.from_pretrained("AnkitNayak-dev/QJev-1.5B")
# Or 4-bit quantized: model = QJev.from_pretrained("AnkitNayak-dev/QJev-1.5B-4bit")
# Or PEFT adapter: model = QJev.from_pretrained("AnkitNayak-dev/QJev-1.5B-LoRA")
# Execute decision inference
result = model.decide(
context="Production database primary replica CPU utilization reached 100% and connection pool is exhausted.",
choices=[
"Trigger immediate automated failover to standby replica",
"Restart the primary database daemon immediately",
"Double connection pool limits without scaling compute",
"Silence alerts and monitor for next 15 minutes"
]
)
print(f"Recommended Action: {result['best_choice']} ({result['confidence']}%)")
for choice, prob in result['probabilities'].items():
print(f" [{prob:>5.2f}%] {choice}")
Recommended Action: Trigger immediate automated failover to standby replica (88.28%)
[88.28%] Trigger immediate automated failover to standby replica
[ 6.54%] Restart the primary database daemon immediately
[ 4.79%] Double connection pool limits without scaling compute
[ 0.45%] Silence alerts and monitor for next 15 minutes
python qjev.py --context "Patient presents with acute chest pain and elevated troponin levels." --choices "Emergency PCI" "Aspirin" "Observation"
serve.py)A production FastAPI service providing REST, Jev protocol, and OpenAI-compatible endpoints:
python serve.py --model AnkitNayak-dev/QJev-1.5B --port 8000
POST /v1/decide - Direct decision endpoint returning structured probabilities.POST /v1/systemone - Official BenchmarkHeaven JEV evaluation protocol.POST /v1/chat/completions - Drop-in OpenAI-compatible wrapper.curl -X POST http://localhost:8000/v1/decide \
-H "Content-Type: application/json" \
-d '{
"context": "Customer requests immediate refund after policy window has expired.",
"options": ["Approve refund", "Issue store credit", "Deny request"]
}'
A preconfigured Modelfile is provided in the repository root:
ollama create qjev-1.5b -f Modelfile
ollama run qjev-1.5b
Querying via the Python SDK:
from qjev import QJev
client = QJev.from_ollama("http://localhost:11434", model="qjev-1.5b")
result = client.decide("Route network traffic to cluster Alpha or Beta?", ["Alpha", "Beta"])
Convert weights to GGUF format for edge CPU and local deployment:
python scripts/export_gguf.py --model-dir export/QJev-1.5B-Phase3 --out-file qjev-1.5b-f16.gguf
./llama-quantize qjev-1.5b-f16.gguf qjev-1.5b-q4_k_m.gguf Q4_K_M
vllm serve AnkitNayak-dev/QJev-1.5B --task classify --port 8000
[Prompt Context + Formatted Candidate Choices]
β
βΌ
ββββββββββββββββββββββββββ
β Qwen2.5-1.5B Backbone β (QLoRA adapted: q, k, v, o projections)
βββββββββββββ¬βββββββββββββ
β (Final Hidden State: 1536 dim)
βΌ
ββββββββββββββββββββββββββ
β LayerNorm(1536) β
βββββββββββββ¬βββββββββββββ
β
βΌ
ββββββββββββββββββββββββββ
β Linear(1536 -> 512) β (GELU Activation, Dropout 0.1)
βββββββββββββ¬βββββββββββββ
β
βΌ
ββββββββββββββββββββββββββ
β Linear(512 -> 10) β (Dynamic Logit Masking for N Choices)
βββββββββββββ¬βββββββββββββ
β
βΌ
[Calibrated Softmax Probabilities]
Qwen/Qwen2.5-1.5B-InstructSargeDev/jev-distill-corpus-v3 (655,806 distillation examples)βββ README.md # Project documentation and benchmarks
βββ LICENSE # Apache 2.0 open-source license
βββ requirements.txt # Python dependencies
βββ Modelfile # Ollama configuration
βββ qjev.py # Unified Python SDK and CLI client
βββ serve.py # Production FastAPI service
βββ predict.py # Forward pass evaluation script
βββ handler.py # Hugging Face inference handler
β
βββ benchmark/ # Benchmark evaluation suites
β βββ run_jevbench.py # JevBench test harness runner
β βββ self_test_benchmarkheaven.py # Official BenchmarkHeaven evaluation runner
β βββ self_test_phase3.py # Phase 3 verification runner
β βββ self_test_step1.py # Calibration sweep script
β
βββ reports/ # Empirical evaluation logs
β βββ jevbench/ # Benchmark records
β βββ self_test_phase3_consistency.json # Official 62.53% report
β βββ jevbench_phase1_report.json
β βββ jevbench_phase2_report.json
β
βββ src/ # Model definitions and training routines
β βββ model.py # MLP Decision Head and Qwen integration
β βββ dataset.py # Streaming ChatML dataset loader
β βββ train.py # Phase 1 initialization
β βββ train_phase2.py # Phase 2 QLoRA training
β βββ train_phase3_consistency.py # Phase 3 consistency alignment
β βββ evaluate.py # Metric evaluation routines
β
βββ scripts/ # Utilities and deployment tools
βββ build_consistency_dataset.py # Consistency dataset generator
βββ merge_phase3.py # Weight fusing and export
βββ merge_and_export.py # Model export routine
βββ export_gguf.py # GGUF conversion utility
βββ publish_all_hf.py # Hugging Face hub publishing script
βββ download_data.py # Dataset download helper
βββ download_model.py # Base model download helper
@misc{nayak2026qjev15b,
author = {Ankit Nayak},
title = {QJev-1.5B: High-Speed System 1 JEV Decision Model based on Qwen2.5},
year = {2026},
publisher = {GitHub and Hugging Face},
howpublished = {\url{https://github.com/AnkitNayak-dev/QJev-1.5B}}
}
This project is licensed under the Apache 2.0 License.
Base model weights are governed by the Alibaba Cloud Qwen License. Training data is provided by SargeDev (jev-distill-corpus-v3).
model.safetensors 3.1 GBlora_adapter/adapter_model.bin 18 MBlora_adapter/adapter_model.safetensors 17 MBdecision_head.pt 3 MBdecision_head.safetensors 3 MBQwen/Qwen2.5-1.5B-Instruct