← Local AI Release Board

Goekdeniz-Guelmez/Omoikane-50M-PoC

archived · low-signalmodel

Omoikane (思兼, the Shinto god of wisdom and thought, "the one who combines many thoughts") is a small hybrid language model built and trained end to end on a single Mac mini (Apple M4, 32 GB unified memory, no CUDA). Dense, 49M params. Context: 4K tokens (config). License: mit. Vendor-reported: "The MTP head predicts token t+2 with 26.3% top-1 accuracy (the main head predicts t+1 with 31.1%)."

Org
Goekdeniz-Guelmez
Released
2026-11-12
Status
archived · low-signal
License
mit (permissive)
Parameters
49M
Architecture
dense · OmoikaneForCausalLM
Context
4,096 tokens (config.json)
Task
text-generation · text
Signal
3 HF likes · 24 downloads
Archived
2026-11-16 (no longer listed by the collector; stats as last seen)

Runs on

estimate from params; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.

Links

First published here 2026-11-12 · updated 2026-11-16 · markdown · JSON