Neutrino-1 8B

Announcement

Introducing the Neutrino-1 models

Three models in a proprietary ternary-family format, served by one engine from one artifact on datacenter GPUs, Apple silicon, and desktop CPUs.

Neutrino-1 8B is an 8.19B-parameter decoder-only transformer that ships as one 3.88 GB file. Every one of its 252 transformer linears is stored in a proprietary ternary-family weight format eight times smaller than fp16; the weights stay bit-packed at rest and are decoded inside the matrix kernels, so nothing in the decode path is stored as fp16 or fp32 weight material.

Small weights change the serving economics. Single-stream decode is bound by how many bytes move per token, so a 3.88 GB working set decodes at rates a 16 GB fp16 artifact cannot reach on the same memory system, and the whole model fits beside its KV cache on an 8 GB GPU or a 16 GB laptop. The same container serves every platform below without conversion.

A dense decoder-only transformer. Grouped-query attention holds the KV cache at a quarter of the query width, 144 KiB per token at fp16, so a 4k-token session costs 0.60 GB of cache beside the 3.88 GB of weights.

Only the transformer linears carry the coded format. The two embedding tensors stay int8 because their rows are read one token at a time, not multiplied against the full activation stream, and the normalization weights are too small to be worth coding. A third of the file is vocabulary.

Across the 6.95B coded weights, 62.63% sit at zero and the remainder splits 18.68% plus to 18.69% minus: sign-balanced to a hundredth of a point with no constraint asking for it. The balance is not uniform in depth. The gate and down feed-forward projections spike to 70 to 72% zeros in layers 1 through 3 while all four attention projections hold within about one point of 62% at every depth: the early feed-forward blocks shed weights the network does not need, and attention keeps a constant code density from layer 0 to layer 35.

One container, three doors. The download is a coded transport of the container, not a compressed copy of an fp16 model: it expands bit-exactly to the file every runtime executes, and that one file is what runs on a datacenter GPU and on a laptop alike.

24.9 tok/s on an Apple M5, CPU only, 9 threads

The one-command door: pip install fermion-research downloads the container and the platform-matching native binary. CPU runtimes for macOS arm64 and Linux x86-64, with a bit-exact torch reference path underneath.

30.7 tok/s on an NVIDIA L4, 4.68 GiB at 4k context

The llama.cpp door: the container converted to GGUF with our weight types, loaded by our public llama.cpp fork. Runs llama-completion and llama-bench, with full CUDA offload.

33.7 tok/s on a base M5 MacBook

The Apple-silicon door: Python-native runtime with custom Metal kernels. The container is memory-mapped and the packed planes are decoded inside the GEMV kernels.

Neutrino-1 0.6B drafts a run of tokens, the 8B scores the whole run in one forward pass, and the agreeing prefix is kept. A draft token is accepted only when it equals the 8B's own argmax, so the output stream is the plain greedy stream: on the shipping configuration, 27,648 consecutive tokens matched with zero divergences.

Both containers are the same format and run on the same binaries, so the draft loads into the verifier's own process with no second deployment and no conversion step. Its 328 MB sit beside the 8B's 3.88 GB, and 4k tokens of shared context cost exactly one gibibyte of cache across the pair.

Open weights under the Apache License 2.0. Commercial use, modification, fine-tuning, and redistribution are permitted, with no access request and no acceptance form. The model is a derivative of Qwen3-8B, itself Apache-2.0.