Dev Builds » 20260726-1046

You are viewing an old NCM Stockfish dev build test. You may find the most recent dev build tests using Stockfish 15 as the baseline here.

Use this dev build

NCM plays each Stockfish dev build 20,000 times against Stockfish 14. This yields an approximate Elo difference and establishes confidence in the strength of the dev builds.

Summary

Host	Duration	Avg Base NPS	Games	WLD	Standard Elo	Ptnml(0-2)	Gamepair Elo

Test Detail

ID	Host	Base NPS	Games	WLD	Standard Elo	Ptnml(0-2)	Gamepair Elo	CLI	PGN

Commit

Commit ID	c8b91a2b9a770f3e9bddbe7863fcf2358d462f05
Author	tony
Date	2026-07-26 10:46:11 UTC
Update main net to nn-ab28990d4ea3.nnue Passed STC https://tests.stockfishchess.org/tests/view/6a5a05955529b8472df811c6 ``` LLR: 2.94 (-2.94,2.94) <0.00,2.00> Total: 35520 W: 9419 L: 9104 D: 16997 Ptnml(0-2): 103, 4119, 9016, 4404, 118 ``` Passed LTC https://tests.stockfishchess.org/tests/view/6a5a4b3a5529b8472df81254 ``` LLR: 2.95 (-2.94,2.94) <0.50,2.50> Total: 97590 W: 25549 L: 25108 D: 46933 Ptnml(0-2): 60, 10413, 27405, 10860, 57 ``` Trained net with slightly optimized recipe. Instead of completely changing the lambda for the last stages, instead use a smooth transition following a cosine schedule. Recipe can be found here https://github.com/vondele/nettest/pull/426 ~~Slight regression in classic matetrack that is currently being investigated.~~ Seems like the regression was only there before rebasing... Unclear what caused it exactly closes https://github.com/official-stockfish/Stockfish/pull/6993 Bench: 2467577