Text Classification
Transformers
Safetensors
PyTorch
qwen2
text-generation
causal-lm
qwen
decision-making
system-one
reasoning
jev
jevbench
joint-expected-value
classification
qlora
agent
decision-model
fast-inference
text-embeddings-inference
Instructions to use devankit7873/QJev-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use devankit7873/QJev-1.5B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="devankit7873/QJev-1.5B")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("devankit7873/QJev-1.5B") model = AutoModelForCausalLM.from_pretrained("devankit7873/QJev-1.5B", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 14,921 Bytes
f43096a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 | ---
license: apache-2.0
base_model: Qwen/Qwen2.5-1.5B-Instruct
library_name: transformers
tags:
- transformers
- pytorch
- causal-lm
- qwen
- qwen2
- decision-making
- system-one
- reasoning
- jev
- jevbench
- joint-expected-value
- classification
- safetensors
- qlora
- agent
- decision-model
- text-classification
- fast-inference
datasets:
- SargeDev/jev-distill-corpus-v3
metrics:
- accuracy
- tvd
- kl_divergence
pipeline_tag: text-classification
---
<div align="center">
# β‘ QJev-1.5B: High-Speed System 1 JEV Decision Model
### Single-Pass Real-Time Decision Engine (~31 ms Forward Pass β’ Ranked #1 Sub-2B Model)
[](https://huggingface.co/devankit7873/QJev-1.5B)
[](https://github.com/AnkitNayak-dev/QJev-1.5B)
[-8A2BE2?style=for-the-badge)](https://benchmarkheaven.com/jev-models)
[-blue?style=for-the-badge)](https://benchmarkheaven.com/jev-models)
[](LICENSE)
---
### π³ Hugging Face Model Tree & Flavors:
```
Qwen/Qwen2.5-1.5B-Instruct (Root Base)
βββ β‘ devankit7873/QJev-1.5B (Flagship Fused 16-bit β’ 3.08 GB β’ Zero PEFT dependency)
βββ π devankit7873/QJev-1.5B-LoRA (PEFT Adapter β’ 20 MB β’ Modular edge loading)
βββ ποΈ devankit7873/QJev-1.5B-4bit (NF4 Quantized β’ 1.15 GB β’ 1.07 GB VRAM)
```
[News](#-news) β’ [Overview](#-overview-system-1-decision-paradigm) β’ [Leaderboard & Benchmarks](#-official-benchmarkheaven--jevbench-results) β’ [Quickstart](#-quickstart--usage) β’ [Multi-Engine Serving](#-multi-engine-inference-deployment) β’ [Architecture](#-model-architecture) β’ [Repo Structure](#-repository-structure) β’ [Citation](#-citation)
</div>
---
## π’ News
* **[2026-11-17] π― Phase 3 Consistency Release & Multi-Engine Serving:**
- Trained with cross-format symmetric consistency loss, boosting official **JevBench strict tolerance (0.05) score from 58.67% to 62.53%** (+3.86% overall, +15.2% on choice conditioning).
- Officially ranks as the **#1 decision model in the sub-2B tier**, outperforming multiple 4B baselines on [BenchmarkHeaven](https://benchmarkheaven.com/jev-models).
- Shipped native deployment support for **vLLM**, **Ollama (`Modelfile`)**, **llama.cpp (GGUF)**, and **production FastAPI (`serve.py`)**.
* **[2026-11-16] π Initial Release:**
- Full QLoRA joint training across all 655,806 distillation samples of `SargeDev/jev-distill-corpus-v3`.
- Achieves **71.50% relaxed metamorphic score** in **3.4 GB VRAM at ~31 ms per decision**.
---
## π‘ Overview: System 1 Decision Paradigm
Traditional LLMs make decisions through slow, non-deterministic token-by-token autoregressive generation (e.g. generating `"Option A"` over 300β600 ms).
**`QJev-1.5B`** introduces a **System 1 Decision Architecture**:
1. Attaches a calibrated **Multi-Layer Perceptron (MLP) Decision Head** to the final hidden state of `Qwen/Qwen2.5-1.5B-Instruct`.
2. Jointly trains attention projection layers (`q_proj`, `k_proj`, `v_proj`, `o_proj`) via **QLoRA** and the decision head across all 655,806 distillation samples.
3. In Phase 3, fine-tunes with soft target KL-divergence consistency loss across multiple-choice, boolean probes, and permutation groups.
4. Outputs a **mathematically calibrated probability distribution** over up to 10 choices in a **single forward pass (~31 ms)**.
---
## π Official BenchmarkHeaven & JevBench Results
Evaluated against the official **JevBench** test harness (`jevbench-mini`, 240 cases, 50 scored metamorphic relations, 1,468 trials):
### 1. Leaderboard Placement (BenchmarkHeaven JEV Leaderboard)
Under the official BenchmarkHeaven standard (**strict tolerance = 0.05**):
| Model | Parameters | Benchmark Score (tol=0.05) | Rank / Category | Latency | VRAM 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%** | **π #1 Sub-2B (Beats 4B)** | **~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% | Server Tier | ~220 ms | ~55.0 GB |
> **Note on Evaluation Tolerances:**
> - **Official Strict (`tol = 0.05`):** **62.53%** (Official score reported on BenchmarkHeaven public leaderboard).
> - **Relaxed Exploration (`tol = 0.10`):** **71.50%** (Lenient consistency threshold).
### 2. Breakdown Across Metamorphic Dimensions
| Dimension | Baseline (Pre-Consistency) | QJev-1.5B (Phase 3 Consistency) | Gain | Practical Meaning |
| :--- | :---: | :---: | :---: | :--- |
| **`BAT.isolation`** | **100.0%** | **100.0%** | **0.0%** | Batching independence (Zero cross-talk between batch items) |
| **`REP.type`** | 77.1% | **83.3%** | **+6.2%** | Format coherence across multi-choice, boolean, and ordinal prompts |
| **`REP.invariance`** | 73.2% | **74.4%** | **+1.2%** | Immune to choice reordering and prompt syntax perturbations |
| **`LOG.bounds`** | 62.5% | **67.9%** | **+5.4%** | Strict adherence to probabilistic upper and lower bounds |
| **`CHO.conditioning`** | 41.7% | **56.9%** | **+15.2%** | Removing irrelevant choices preserves odds ratios (Independence of Irrelevant Alternatives) |
| **`REP.equivariance`** | 48.6% | **55.6%** | **+7.0%** | Symmetrical probability shifts under negation transformations |
| **`MEA.partition`** | 30.7% | **41.1%** | **+10.4%** | Probability mass preservation across option clustering |
| **OVERALL (Strict 0.05)** | **58.67%** | **62.53%** | **+3.86%** | **Official Benchmark Standard** |
Machine-readable evaluation reports are archived in `reports/jevbench/`.
---
## π Quickstart & Usage
### 1. Installation
```bash
git clone https://github.com/AnkitNayak-dev/QJev-1.5B.git
cd QJev-1.5B
pip install -r requirements.txt
```
### 2. Fast Python API (3 Lines of Code)
```python
from qjev import QJev
# Load ANY edition (Full Standalone, 4-bit Quantized, or LoRA Adapter) in 1 line:
model = QJev.from_pretrained("devankit7873/QJev-1.5B")
# Or 4-bit (1.07 GB VRAM): model = QJev.from_pretrained("devankit7873/QJev-1.5B-4bit")
# Or LoRA (20 MB download): model = QJev.from_pretrained("devankit7873/QJev-1.5B-LoRA")
# 1-Line Decision (~31 ms)
result = model.decide(
context="Production database primary replica CPU spiked to 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"π Recommended Action: {result['best_choice']} ({result['confidence']}%)")
for choice, prob in result['probabilities'].items():
print(f" [{prob:>5.2f}%] {choice}")
```
**Expected Output:**
```text
π 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
```
### 3. CLI Command-Line Decision
```bash
python qjev.py --context "Patient has severe chest pain and elevated troponin" --choices "Emergency PCI" "Aspirin" "Wait"
```
---
## π Multi-Engine Inference Deployment
QJev-1.5B natively supports **all standard LLM serving runtimes**:
### 1. High-Performance Microservice Server (`serve.py`)
Run the unified FastAPI server exposing REST, System One, and OpenAI compatible endpoints:
```bash
python serve.py --model devankit7873/QJev-1.5B --port 8000
```
Endpoints provided:
* **REST Decision API:** `POST /v1/decide`
* **Official Jev Protocol (BenchmarkHeaven):** `POST /v1/systemone`
* **OpenAI Drop-In Compatible:** `POST /v1/chat/completions`
Query via cURL:
```bash
curl -X POST http://localhost:8000/v1/decide \
-H "Content-Type: application/json" \
-d '{
"context": "Customer requests immediate refund after 45 days.",
"options": ["Approve refund", "Offer store credit", "Deny request"]
}'
```
### 2. Ollama (Local Mac / PC / Linux)
Package and run locally with Ollama using the included `Modelfile`:
```bash
ollama create qjev-1.5b -f Modelfile
ollama run qjev-1.5b
```
Or query from Python via `QJev`:
```python
from qjev import QJev
model = QJev.from_ollama("http://localhost:11434", model="qjev-1.5b")
result = model.decide("Route packet to cluster A or B?", ["Cluster A", "Cluster B"])
```
### 3. llama.cpp / GGUF (LM Studio & Edge CPU)
Export to GGUF format for edge devices using the export tool:
```bash
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
```
### 4. vLLM (Production Continuous Batching)
```bash
vllm serve devankit7873/QJev-1.5B --task classify --port 8000
```
---
## ποΈ Model Architecture
```
[User Context + Formatted Choices]
β
βΌ
ββββββββββββββββββββββββββ
β Qwen2.5-1.5B Backbone β (QLoRA adapted on q, k, v, o projections)
βββββββββββββ¬βββββββββββββ
β (Last Hidden State: 1536 dim)
βΌ
ββββββββββββββββββββββββββ
β LayerNorm(1536) β
βββββββββββββ¬βββββββββββββ
β
βΌ
ββββββββββββββββββββββββββ
β Linear(1536 -> 512) β (GELU + Dropout 0.1)
βββββββββββββ¬βββββββββββββ
β
βΌ
ββββββββββββββββββββββββββ
β Linear(512 -> 10) β (Dynamic Choice Masking)
βββββββββββββ¬βββββββββββββ
β
βΌ
[Calibrated Softmax Probabilities]
```
---
## π Repository Structure
```text
βββ README.md # Project showcase, benchmarks, and documentation
βββ LICENSE # Apache 2.0 open-source license
βββ requirements.txt # Python dependencies
βββ Modelfile # Ollama build definition
βββ qjev.py # Universal Python SDK & CLI client
βββ serve.py # Production FastAPI server (REST, Jev, OpenAI)
βββ predict.py # Single-pass forward evaluation demo
βββ handler.py # Hugging Face Inference Endpoint handler
β
βββ benchmark/ # JevBench self-test and evaluation suites
β βββ run_jevbench.py # JevBench metamorphic test harness
β βββ self_test_benchmarkheaven.py # Official BenchmarkHeaven standard runner
β βββ self_test_phase3.py # Phase 3 consistency verification runner
β βββ self_test_step1.py # Calibrated temperature sweep suite
β
βββ reports/ # Empirical evaluation records
β βββ jevbench/ # JSON benchmark results across phases
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/ # Core modeling and training code
β βββ model.py # MLP Decision Head + Qwen architecture
β βββ dataset.py # Streaming ChatML dataset builder
β βββ train.py # Phase 1 baseline training
β βββ train_phase2.py # Phase 2 QLoRA joint training (655k samples)
β βββ train_phase3_consistency.py # Phase 3 KL consistency alignment
β βββ evaluate.py # Out-of-Distribution evaluation metrics
β
βββ scripts/ # Engineering and deployment utilities
βββ build_consistency_dataset.py # 20k paired consistency dataset generator
βββ merge_phase3.py # Phase 3 LoRA fusing & packaging
βββ merge_and_export.py # Base export utility
βββ export_gguf.py # GGUF converter for llama.cpp / LM Studio
βββ publish_all_hf.py # Hugging Face Hub publisher
βββ download_data.py # Dataset downloader
βββ download_model.py # Base model downloader
```
---
## π» Hardware & Training Infrastructure
- **Hardware:** HP OMEN Laptop (NVIDIA GeForce RTX 5060 Laptop GPU, 8GB VRAM)
- **Host OS:** Windows 11 Pro 64-bit
- **CUDA / Driver:** CUDA 12.8 / Driver 621.24
- **Precision:** Float32 Matmul Precision (`high`) with Tensor Core TF32 acceleration
- **Training Trajectory:**
- **Phase 1:** MLP Head alignment on frozen backbone (1.5 hrs).
- **Phase 2:** Joint QLoRA + MLP training across 655,806 samples (18.5 hrs, strictly 5.9 GB VRAM).
- **Phase 3:** Paired consistency KL alignment (27 mins, 1,250 steps, batch size 16).
---
## π Citation
If you use **QJev-1.5B** or its benchmark methodology, please cite:
```bibtex
@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}}
}
```
---
## π License
This model and codebase are released under the **Apache 2.0 License**.
Base weights are governed by Alibaba Cloud Qwen License. Distillation corpus is provided by SargeDev (`jev-distill-corpus-v3`).
|