EXP-9 / completed
State-conditioned system evaluation
CM-6 / completed system evaluation
Read experiment →CM-6 / research checkpoint
State-conditioned sanity checkpoint
Warm start from CM-5 with expanded vocabulary and new input/objective; not promoted.
86,235,648 parameters. 120-slot prefix encoding current board/state plus optional 16-ply history; up to 122 inference positions, with 512 allocated vocabulary IDs. Complete past game history is not required, and repetition history may be unknown. Different training objective; loss is not comparable to full-history runs. No reasoning traces.
Checkpoint SHA-256 7b63bab559741893b632f456e5f9a8e48b15e48fc9c47c302734da9d613dfe85
Initialized from CM-5.
Experiments
Inspect and reproduce
Recorded evidence and source · All versioned source, including Rust · Published downloads and limitations · Games and checkpoint filters
Research-only; not a promotion candidate. The one-hour training run validates the state-input pipeline. EXP-9 completed 100 games: 13W / 7D / 80L, no unfinished games (16.5% score). This is a system-level result, not an isolated model comparison.