| Version | Release Date | CCRL 40/15 | CCRL Blitz | CEGT 40/20 | ipman r9 list | Chess324 Top15 | SP-CC Top 15 |
|---|---|---|---|---|---|---|---|
| 1.9 | 2026-01-04 | 3587 (#10) | 3727 (#13) | 3569 (#11) | 3537 (#20**) | 3734 (#8) | 3680 (#14) |
| 1.8 | 2025-07-22 | 3564 (#20) | 3697 (#16) | 3544 (#18) | 3517 | 3680 (#12) | |
| 1.7 | 2025-05-31 | 3528 | 3642 | 3491 | 3448 | ||
| 1.6.1 | 2025-05-15 | 3500* | 3622 | 3440* | |||
| 1.6 | 2025-04-27 | 3490 | 3600* | 3433 | |||
| 1.5 | 2025-04-18 | 3450* | 3500 | 3350 | |||
| 1.4.1 | 2025-04-05 | 3425 | 3450* | ||||
| 1.3.1415 | 2025-03-14 | 3365 | 3401 | ||||
| 1.3 | 2025-03-07 | 3201 | |||||
| 1.2 | 2025-02-21 | 3120 | |||||
| 1.1 | 2025-01-24 | 2432 | |||||
| 1.0 | 2025-01-20 | 2100* |
**#12 if you only keep 1 version per engine
Supports FRC, also known as Chess960
The search is a standard alpha-beta search with many enhancements.
The evaluation is done using a neural net trained on self play games from zero knowledge, as well as games played by Vine, the MCTS Chess engine that @aronpetko and I have been working on. The networks are trained using the excellent open source bullet neural network trainer. For specifics about the network please see src/nnue.zig.
- Get the network with
git submodule update --init --depth 1 - Install zig (0.15.2)
zig build --release=fast --prefix <installation path>(for example--prefix ~/.localwill put pawnocchio in~/.local/bin/pawnocchio) The Makefile is only intended to be used for testing on Openbench.
- The code is licensed under the GPLv3 license. Full text can be found in LICENSE in the project root
- The assets are licensed under the CC-BY-ND 4.0 license. Full text can be found in assets/LICENSE
- Pyrrhic by Andrew Grant for tablebase probing, under the MIT license.
- src/bounded_array.zig from the zig standard library under the MIT License with some minor modifications.
- Jackal by snekkers for inspiration for styling this readme
