YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
kotodama-3b-posttrain-archive
Final checkpoints of the kotodama 3B post-training experiments (2026-06 .. 2026-09): DPO rounds and variants (dpo/), SFT sweeps and gates (sft-*/), RLVR binding rounds (rlvr-binding/), and the judge student (judge_student/). One model-only bf16 checkpoint per run, plus small run configs/metrics where present. Base: aethera-gp/kotodama-3b-base-final. Crowned models: aethera-gp/kotodama-3b-releases. Load with kotodama.ckpt.load (github.com/LuxiaSL/kotodama).
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support