You can deploy this model and start earning money today!

model-rampage/
Qwen3-1.7B-mlx-8bit

Name: Qwen3-1.7B-mlx-8bit
Rating: 2.5 (5 reviews)

Views

No views yet

Qwen/Qwen3-1.7B (8-bit Native MLX Family Core)

This asset is an optimized execution variant of Qwen/Qwen3-1.7B, compiled natively for Apple Silicon hardware acceleration via the MLX framework under the Model Rampage Telemetry Engine.

📊 Cross-Precision Family Matrix

The comprehensive family optimization footprint table generated via our open control evaluation metrics.

Precision Target	Disk Footprint	Peak VRAM	Quantized PPL	Brain Fidelity	Active Selection
2-bit Precision	0.51 GB	0.57 GB	639987.3113	0.0%
3-bit Precision	0.71 GB	0.77 GB	50.5746	37.95%
4-bit Precision	0.91 GB	0.97 GB	24.0658	79.75%
8-bit Precision	1.71 GB	1.77 GB	19.3544	99.16%	👈 (This Repo)

Baseline Reference Model Perplexity Control: 19.1917
Standardized Evaluation Corpus Suite: Salesforce/WikiText-2 (Raw Validation Split)
Data Context Coverage Scope: Full Dataset (212 Chunks)

💻 Cross-Device Throughput Estimation Map

Projected prompt throughput estimations generated via empirical memory-bandwidth scaling configurations relative to host benchmark runtime execution.

Quant Format	🟢 M-Series Base (~100 GB/s)	🔵 M-Series Pro (~150-200 GB/s)	🟣 M-Series Max (~300-400 GB/s)	🔥 M-Series Ultra (~800 GB/s)
2-bit	56.79 tok/s	103.25 tok/s	201.34 tok/s	351.05 tok/s
3-bit	55.61 tok/s	101.11 tok/s	197.16 tok/s	343.77 tok/s
4-bit	59.06 tok/s	107.38 tok/s	209.39 tok/s	365.09 tok/s
8-bit	41.26 tok/s	75.02 tok/s	146.29 tok/s	255.07 tok/s	👈

🚀 Native Implementation Scaffolding

python

from mlx_lm import load, generate
model, tokenizer = load('model-rampage/Qwen3-1.7B-mlx-8bit')

response = generate(
    model, 
    tokenizer, 
    prompt="Your system prompt execution sequence goes here.", 
    max_tokens=128, 
    verbose=True
)

Telemetry tracked via Model Rampage Engine — Control Verification Core.