News
2026.10.21: Boris-2-Preview-2
The second preview of Boris-2: the main run's 100B-token checkpoint with the same 2.5B-token learning-rate decay applied as Preview-1. Every benchmark moved up — HellaSwag 46.60, ARC-Easy 59.13, ARC-Challenge 30.20, PIQA 70.84 — and ArithMark-3 rose 4.6 points to 42.20 on grade-school word-problem data added to the decay, taking the Open SLM Index from 27.91 to 30.58. The ArithMark gain does not yet carry over to SVAMP or ASDiv, which stayed level with the pilot baseline. The decay start and length were fixed in advance, and no checkpoint was picked on its scores. See the Boris page.
2026.10.09: Boris-2-Preview-1
A first look at Boris-2, a 143.7M-parameter base model trained from
scratch on a single NVIDIA V100 32GB, on a run that continues to 200B
tokens. Preview-1 is its 50B-token checkpoint with a 2.5B-token decay
applied. The architecture is new for the family: 31 layers at d=512, causal
Canon convolutions on the attention projections, a 15M-parameter exact
n-gram table carried over from the Boris-1.7 experiment, a 2048-token
context, and its own 32,768-entry byte-level BPE. Open SLM Index 27.91,
ahead of SmolLM2-135M's 27.13 on the leaderboard's published figures, after
52.5B tokens against SmolLM2's 2T. It loads through transformers
with trust_remote_code=True.
2026.08.31: Boris-1.7-D60M-n30M
A 60M-parameter dense decoder-only trunk (10 layers, d=768, GQA, SwiGLU,
RoPE, RMSNorm) with a 30M-parameter n-gram embedding table spliced into the
residual stream after block 2, trained on 2.5B tokens in 22h 37m. One
experiment, one variable: does 30M of lookup buy what 30M of dense weight
would at a fixed token budget? The table is exact rather than hashed
— 281,250 bigram and 187,500 trigram rows plus a shared OOV row, one
searchsorted gather per position, no matmul. Ablating it at
inference costs 0.069 bits/char and 0.7–3.1 points across six
benchmarks. It also ships its own 32,768-entry byte-level BPE trained on the
corpus, so it is not a transformers checkpoint and loads via the
loader in its repository. See the Boris page.
2026.08.12: Lucy-2
A 112.5M-parameter text-to-image diffusion transformer pretrained from scratch at 128×128 — rectified flow, DiT backbone, frozen VAE and CLIP text encoder, roughly 44 GPU-hours across a CC3M pretrain and two finetuning stages. Released for research use only, because CelebA is licensed that way and CelebA face crops are what finally fixed faces. Full write-up, including the mandatory negative prompt, on the Lucy page.
2026.08.09: Lucy
A 71.4M-parameter unconditional DDPM generating 128×128 faces, trained from scratch on CelebA for 60,000 steps in 9.54 hours. The released weights are the EMA weights, and the repository carries the training script, the exact launch command, the complete run log, and a fixed-seed EMA preview grid every 3,000 steps.
2026.08.07: littlerock-1M
littlerock-1M, a 1,002,720-parameter Llama-architecture model trained from
scratch on 5B tokens (40% FineWeb-Edu, 60% DCLM-baseline) in 1 hour 49
minutes. It is published alongside littlerock-1M-arithmax, an
arithmetic finetune released as a research artifact rather than a model to
use.
2026.08.01: Boris-1.3-125M-Instruct and GGUF builds
Instruction-tuned on smol-smoltalk and OASST1 over 5.6 hours, with quantized GGUF builds for llama.cpp published the same day.
2026.07.31: Boris-1.3-125M
Seven continued-pretraining passes on the Boris-125M base checkpoint, adding ~2.66B tokens — about 106% more than the original 2.50B-token pretraining run — and raising Mean-6 from 37.88 to 39.38. Leading with FineWeb-Edu before DCLM avoided the ARC regression seen at 75M.
2026.07.26: Boris-1.3-75M and Instruct
Three continued-pretraining passes bring the 75M model to ~3.95B total training tokens and a Mean-6 of 36.79, with instruct and GGUF variants alongside.
2026.07.23: Boris-250M, and GGUF builds across the board
The largest Boris to date, in base and instruct form, plus quantized GGUF builds for the 75M, 125M, and 250M instruct models.
2026.07.18: Boris-75M and Boris-75M-Instruct
A 77.4M-parameter sibling to the original release.
2026.07.15: NMAI becomes OpenCerebral; first release
New Millennium Artificial Intelligence, a division of the Klondike Software Project, is renamed OpenCerebral. The organization, models, and maintainers are unchanged. The same day, Boris-125M and Boris-125M-Instruct are published as the project's first public models.
References to NMAI and to the old KSP-NMAI repository paths
— which still appear in some model cards — refer to
OpenCerebral.