JevK5-chat-mlx-4bit

JevK5-mlx-4bit#

4-bit MLX conversion of alibiserikbay/JevK5 optimized for Apple Silicon native GPU inference.

Converted by: SirSahOl Source Model: alibiserikbay/JevK5 Framework: MLX by Apple Quantization: 4-bit (Average 4.50 bits per weight) Format: safetensors License: apache-2.0

Model Details#

• Architecture: Qwen3_5ForCausalLM
• Parameters: ~4.2B
• Context Length: 262,144 tokens
• Format: MLX (Apple Silicon native GPU format)
• Quantization: 4-bit (Average 4.50 bits per weight)
• Active VRAM Footprint: ~2.7 GB (Minimum recommended: 8 GB Unified Memory)

Quick Start#

Installation#

pip install mlx-lm

Usage#

CLI#

# Chat interactively
mlx_lm.chat --model SirSahOl/JevK5-chat-mlx-4bit

# Generate text
mlx_lm.generate --model SirSahOl/JevK5-chat-mlx-4bit --prompt "Write a short poem about artificial intelligence."

Python API (with Chat Template)#

from mlx_lm import load, generate

model, tokenizer = load("SirSahOl/JevK5-chat-mlx-4bit")

messages = [
    {"role": "user", "content": "Explain quantum superposition in simple terms."}
]
prompt = tokenizer.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)

response = generate(model, tokenizer, prompt=prompt, verbose=True)
print(response)

Performance Benchmarks#

Measured Benchmarks (Apple M1)#

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

Multi-Quantization Comparison#

Evaluate your hardware budget and choose the optimal precision:

Who Should Use This?#

General guidance:

• Use 4-bit if you want to run this model alongside IDEs, browsers, and background development tools.
• Use 8-bit if you have 16GB+ unified memory and require superior reasoning and code accuracy.
• Use 16-bit for research, benchmarking, evaluation, or high-end workstation deployments.

Other Quantization Variants#

LM Studio & Local Inference Setup Guide#

To prevent runaway loops and ensure correct conversational turn-taking, configure custom stop tokens in your local inference runtime:

1. <|im_start|>
2. <|im_end|>
3. <|endoftext|>

Prompt Template Formatting#

• System Prefix: <|im_start|>system\n
• System Suffix: <|im_end|>\n
• User Prefix: <|im_start|>user\n
• Assistant Suffix: <|im_end|>\n<|im_start|>assistant\n

Ollama Quickstart#

FROM SirSahOl/JevK5-chat-mlx-4bit
PARAMETER stop "<|im_start|>"
PARAMETER stop "<|im_end|>"
PARAMETER stop "<|endoftext|>"
PARAMETER temperature 0.7

ollama create jevk5-chat-mlx-4bit -f Modelfile
ollama run jevk5-chat-mlx-4bit

Conversion Details#

Reproduction#

To reproduce this conversion:

pip install mlx-lm==0.31.3
python3 -m mlx_lm.convert --hf-path alibiserikbay/JevK5 --mlx-path output/JevK5-mlx-4bit -q --q-bits 4

Limitations & Known Issues#

• 4-bit group-wise quantization introduces minor precision loss compared to unquantized weights; for deep mathematical derivations or precision-critical reasoning, test the 8-bit or 16-bit variants.
• High context sequences (>32K tokens) require sufficient unified memory headroom; ensure unified memory is not overcommitted.
• This is a weight-only MLX conversion designed specifically for Apple Silicon GPUs (M1/M2/M3/M4 series).

License#

This model conversion inherits the license of the source model: apache-2.0.

See the original model card for full license details.

Changelog#

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