BitNet b1.58 — local AI model by Microsoft
Written by Jakub Rusinowski · Last updated
A research paradigm shift by Microsoft. BitNet b1.58 replaces 16-bit weights with ternary {-1, 0, 1} weights. This eliminates Matrix Multiplication (MatMul) in favor of simple addition, offering extreme speed and efficiency on CPUs.
Variants
The smallest BitNet b1.58 variant needs about 3 GB of VRAM at 1.58-bit — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| BitNet b1.58 3B → 3B | ~2.8 GB | 2,048 | ollama run hf.co/1bitLLM/bitnet_b1_58-3B |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run BitNet b1.58 locally
Install Ollama, then pull the tag.
ollama run hf.co/1bitLLM/bitnet_b1_58-3BPick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: BitNet b1.58 3BRecommended GPU
The cheapest catalogued GPU that runs BitNet b1.58 locally (min 3 GB VRAM) is the Intel Arc B570 (10 GB).
Can I run BitNet b1.58 on my GPU?
BitNet b1.58 — frequently asked questions
How much VRAM does BitNet b1.58 need?
BitNet b1.58 needs about 3 GB VRAM at 1.58-bit quantization for its smallest variant. Variants: BitNet b1.58 3B (3 GB, 1.58-bit). On Apple Silicon, unified memory counts toward this requirement.
Can I run BitNet b1.58 on an RTX 4090 (24 GB)?
Yes — BitNet b1.58 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at 1.58-bit.
What quantization should I use for BitNet b1.58?
Q4_K_M is the best balance of quality and VRAM for BitNet b1.58 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run BitNet b1.58 with Ollama?
Install Ollama, then run: ollama run hf.co/1bitLLM/bitnet_b1_58-3B. This downloads BitNet b1.58 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.