Model Index
Seventeen models across two families, all Apache 2.0, all hosted on Hugging Face. Base models continue text and do not follow instructions. Instruct models are finetuned for conversation. GGUF builds are quantized for llama.cpp.
Boris 1.3 — current
| Model | Type | Params | Released |
|---|---|---|---|
| Boris-1.3-125M | Base | 125M | 2026.07.31 |
| Boris-1.3-125M-Instruct | Instruct | 125M | 2026.08.01 |
| Boris-1.3-125M-Instruct-GGUF | GGUF | 125M | 2026.08.01 |
| Boris-1.3-75M | Base | 77.4M | 2026.07.26 |
| Boris-1.3-75M-Instruct | Instruct | 77.4M | 2026.07.26 |
| Boris-1.3-75M-Instruct-GGUF | GGUF | 77.4M | 2026.07.26 |
Boris 1.0 — original release
| Model | Type | Params | Released |
|---|---|---|---|
| Boris-250M | Base | 250M | 2026.07.23 |
| Boris-250M-Instruct | Instruct | 250M | 2026.07.23 |
| Boris-250M-Instruct-GGUF | GGUF | 250M | 2026.07.23 |
| Boris-125M | Base | 125M | 2026.07.15 |
| Boris-125M-Instruct | Instruct | 125M | 2026.07.15 |
| Boris-125M-Instruct-GGUF | GGUF | 125M | 2026.07.23 |
| Boris-75M | Base | 77.4M | 2026.07.18 |
| Boris-75M-Instruct | Instruct | 77.4M | 2026.07.18 |
| Boris-75M-Instruct-GGUF | GGUF | 77.4M | 2026.07.23 |
littlerock
| Model | Type | Params | Released |
|---|---|---|---|
| littlerock-1M | Base | 1,002,720 | 2026.08.07 |
| littlerock-1M-arithmax | Research artifact | 1,002,720 | 2026.08.07 |
Which one should you use? For conversation at the best quality we currently offer, take Boris-1.3-125M-Instruct. For text continuation or as a finetuning base, take Boris-1.3-125M. For CPU-only or embedded use, take a GGUF build. For scaling research at the smallest end, take littlerock-1M.