BitMamba-2-255M is the ultra-efficient baseline model of the BitMamba-2 family. It integrates 1.58-bit ternary quantization (BitNet) into the Mamba-2 architecture. Despite its small size, it demonstrates stable convergence and surprising reasoning capabilities, serving as the proof-of-concept for scaling ternary State Space Models.
ARM NEON Port — Cross-Platform CPU Inference
An ARM NEON port of the BitMamba-2 inference engine has been developed by Aquantic Research, enabling native inference on Apple Silicon (M1/M2/M3/M4) and ARM-based processors.
Model
Hardware
Speed
Latency/token
RAM
BitMamba-2 255M
Apple M1 (ARM NEON)
82.5 tok/s
12.1 ms
252 MB
BitMamba-2 255M
Intel Core i3-12100F (AVX2)
~146 tok/s
—
252 MB
Key finding: Speed is perfectly constant regardless of sequence length (50, 200, or more tokens). This experimentally validates the O(1) memory property of SSM architectures — unlike Transformers whose memory grows with sequence length.
ARM NEON Port Resources
Code: rasata/bitmamba.cpp — ARM NEON fork with cross-platform dispatch (x86 AVX2 + ARM NEON)
Preprint: "State Space Models as CPU-Native Neural Network Architectures: Experimental Evidence from ARM NEON Inference with 1.58-bit Quantized Mamba" — Gabriel Zo-Hasina Rasatavohary, Aquantic Research, March 2026. To be published on engrXiv (DOI pending).
1# Clone the ARM NEON fork2git clone https://github.com/rasata/bitmamba.cpp
3cd bitmamba.cpp
45# Build (macOS Apple Silicon)6brew install libomp
7cmake -B build && cmake --build build
89# Download weights from this repo10wget https://huggingface.co/rasatavohary/BitMamba-2-0.25B/resolve/main/bitmamba_cpp/bitmamba_255m.bin
1112# Run inference13cd build &&cp../tokenizer.bin .14./bitmamba ../bitmamba_255m.bin "The future of AI is" tokenizer 0.71.10.050.940200
Training Tokens: Trained on high-quality data (FineWeb-Edu, Cosmopedia, Stack-Dedup).
Hardware: Trained on Google Cloud TPU v6e.
📊 Benchmark Results
This model serves as the baseline for our scaling laws analysis.
Benchmark
Metric
BitMamba-2-255M
ARC-Easy
Accuracy
55.51%
PIQA
Accuracy
64.42%
BoolQ
Accuracy
59.30%
HellaSwag
Acc Norm
35.22%
WikiText-2
Perplexity
51.69
As shown in the scaling analysis below, the 255M model (blue line) establishes a stable learning trajectory, which is significantly improved upon by the 1B model (red line).
Scaling Laws
🚀 Usage (Inference)
This model is optimized for extreme edge deployment (IoT, Mobile, Legacy Hardware) using our custom C++ inference engine.
1. Download the Quantized Model
Download the bitmamba_255m.bin file located in the files tab.
