SirSahOl/DeepSeek-R1-Distill-Qwen-1.5B-chat-mlx-16bit

DeepSeek-R1-Distill-Qwen-1.5B-mlx-16bit

16-bit MLX conversion of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B for Apple Silicon.

Converted by: SirSahOl Source model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B Framework: MLX by Apple Quantization: 16-bit Format: safetensors License: mit

---

Quick Start

Installation

pip install mlx-lm

CLI Usage

# Chat interactively
mlx_lm.chat --model SirSahOl/DeepSeek-R1-Distill-Qwen-1.5B-chat-mlx-16bit

# Generate text
mlx_lm.generate --model SirSahOl/DeepSeek-R1-Distill-Qwen-1.5B-chat-mlx-16bit --prompt "Your prompt here"

Python Usage

from mlx_lm import load, generate

model, tokenizer = load("SirSahOl/DeepSeek-R1-Distill-Qwen-1.5B-chat-mlx-16bit")
response = generate(model, tokenizer, prompt="Your prompt here", max_tokens=256)
print(response)

---

Performance Benchmarks

| Metric | 16-bit | |--------|--------|| Tokens/sec | 16.97 | | TTFT | 58.94 ms | | Peak Memory | 280.3 MB |

Benchmarked on Apple M1 with 8GB unified memory. Average over 5 runs with 256 max tokens.

---

Who Should Use This?

Your Hardware	Recommended Quantization
M1/M2 (8GB)	4-bit — Best balance of quality and memory usage
M1/M2 Pro/Max (16-32GB)	8-bit — Higher quality with reasonable memory
M2/M3/M4 Ultra (64GB+)	16-bit — Full precision, no quality loss

General guidance:

• Use 4-bit if you want to run this model alongside other applications
• Use 8-bit if you have the memory and want better quality
• Use 16-bit for research, evaluation, or if memory isn't a concern

---

Other Quantization Variants

Variant	Link
16-bit	SirSahOl/DeepSeek-R1-Distill-Qwen-1.5B-chat-mlx-16bit

---

Conversion Details

Property	Value
Source Model	deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
Quantization	16-bit
mlx-lm Version	0.31.3
Conversion Time	1980.44s
Output Size	3.3 GB
Date	2026-09-11T00:24:16.412595+00:00

Reproduction

To reproduce this conversion:

pip install mlx-lm==0.31.3
python3 -m mlx_lm.convert --hf-path deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --mlx-path output/DeepSeek-R1-Distill-Qwen-1.5B-mlx-16bit

---

Limitations & Known Issues

• Performance may degrade with very long contexts (>8K tokens) at lower quantization levels.
• This is a weight-only conversion; the model architecture and behavior are inherited from the source model.
• Quantization introduces a small quality loss compared to the original model. Lower bit counts = more loss.
• This model requires Apple Silicon (M1 or later) to run with MLX.

---

License

This model conversion inherits the license of the source model: mit.

See the original model card for full license details.

---

Changelog

Version	Date	Changes
v1.0	2026-09-11	Initial conversion

---

Converted with MLX Foundry — a professional pipeline for converting models to Apple MLX format.