I toast to LeelaqueenOdds and its creators. I raise my wine glass high and shout: Long live the queen. We have a queen!!

[Father]

I see that I did not find instructions how to download leelaqueen odds in discord.

searching in google lead me to the following link and I am not sure if it is the LeelaQueenOdds or some old version

https://github.com/notune/LeelaQueenOdds

I see release date is 13.11.2024

Yes, that is the net used by LeelaQueenOdds from last November up until one day ago. We upgraded (?) to a new net also trained for queen odds, by "@Ira_Jed and @Lucario6607. It's not yet clear that it is stronger, but I think on balance the evidence suggests that it is a bit stronger against both human-like bots and against actual humans. Right after the switch, it played four blitz (3'0") games against GM Jaan Ehlvest, who was World number 5 about 35 years ago and is still, at age 63, rated well over the GM threshold at 2532 FIDE. Leela won three and lost one. No games decided by time, but Ehlvest did move a bit faster than necessary for a 3' game, more as if they were 2' games. He said that he thinks this was the main reason for his loss, but even so winning a blitz match at queen odds from a former top-five player indicates that the new net is not bad! By a curious coincidence, Ehlvest is the strongest player I ever defeated in a tournament game, around 1996 I believe, in the World Open.

Good evening, Mr. Larry Kaufman, gentlemen of the Leela Bots team, and Mr. Lichess. I hope you are all well and continue to do well. I wish the best for all of you and for each and every one of you. I also hope that these notes and words reach you, as well as chess enthusiasts and players, and especially those who love challenges against chess computers. Today I want to propose an idea; it is this: "Design some type of monthly open tournament or competition for each of the bots, where humans have the opportunity to play only against the bots and obtain a spot in the competition. This will increase the popularity of the bots and the incentives of the humans; it would also be reflected in the Elo within the bots' table."Sincerely, Pablo

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[pgn][/pgn][/pgn]

[Father]

I see that I did not find instructions how to download leelaqueen odds in discord.

searching in google lead me to the following link and I am not sure if it is the LeelaQueenOdds or some old version

https://github.com/notune/LeelaQueenOdds

I see release date is 13.11.2024

Yes, that is the net used by LeelaQueenOdds from last November up until one day ago. We upgraded (?) to a new net also trained for queen odds, by "@Ira_Jed and @Lucario6607. It's not yet clear that it is stronger, but I think on balance the evidence suggests that it is a bit stronger against both human-like bots and against actual humans. Right after the switch, it played four blitz (3'0") games against GM Jaan Ehlvest, who was World number 5 about 35 years ago and is still, at age 63, rated well over the GM threshold at 2532 FIDE. Leela won three and lost one. No games decided by time, but Ehlvest did move a bit faster than necessary for a 3' game, more as if they were 2' games. He said that he thinks this was the main reason for his loss, but even so winning a blitz match at queen odds from a former top-five player indicates that the new net is not bad! By a curious coincidence, Ehlvest is the strongest player I ever defeated in a tournament game, around 1996 I believe, in the World Open.

Good evening, Mr. Larry Kaufman, gentlemen of the Leela Bots team, and Mr. Lichess. I hope you are all well and continue to do well. I wish the best for all of you and for each and every one of you. I also hope that these notes and words reach you, as well as chess enthusiasts and players, and especially those who love challenges against chess computers. Today I want to propose an idea; it is this: "Design some type of monthly open tournament or competition for each of the bots, where humans have the opportunity to play only against the bots and obtain a spot in the competition. This will increase the popularity of the bots and the incentives of the humans; it would also be reflected in the Elo within the bots' table."Sincerely, Pablo

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[pgn][/pgn][/pgn]

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[/pgn]

[Father]

I see that I did not find instructions how to download leelaqueen odds in discord.

searching in google lead me to the following link and I am not sure if it is the LeelaQueenOdds or some old version

https://github.com/notune/LeelaQueenOdds

I see release date is 13.11.2024

Yes, that is the net used by LeelaQueenOdds from last November up until one day ago. We upgraded (?) to a new net also trained for queen odds, by "@Ira_Jed and @Lucario6607. It's not yet clear that it is stronger, but I think on balance the evidence suggests that it is a bit stronger against both human-like bots and against actual humans. Right after the switch, it played four blitz (3'0") games against GM Jaan Ehlvest, who was World number 5 about 35 years ago and is still, at age 63, rated well over the GM threshold at 2532 FIDE. Leela won three and lost one. No games decided by time, but Ehlvest did move a bit faster than necessary for a 3' game, more as if they were 2' games. He said that he thinks this was the main reason for his loss, but even so winning a blitz match at queen odds from a former top-five player indicates that the new net is not bad! By a curious coincidence, Ehlvest is the strongest player I ever defeated in a tournament game, around 1996 I believe, in the World Open.

Good evening, Mr. Larry Kaufman, gentlemen of the Leela Bots team, and Mr. Lichess. I hope you are all well and continue to do well. I wish the best for all of you and for each and every one of you. I also hope that these notes and words reach you, as well as chess enthusiasts and players, and especially those who love challenges against chess computers. Today I want to propose an idea; it is this: "Design some type of monthly open tournament or competition for each of the bots, where humans have the opportunity to play only against the bots and obtain a spot in the competition. This will increase the popularity of the bots and the incentives of the humans; it would also be reflected in the Elo within the bots' table."Sincerely, Pablo

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[pgn][/pgn][/pgn]

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[/pgn]

Good evening, Mr. Larry Kaufman. Time flies. I'm approaching my first year playing against robots, barely a month away. In this game I'm publishing, I don't understand the robot's 17th move from bishop to king's knight. I'm already facing all four robots; I'll leave the advantage knight for last. I love playing against these wonderful machines.

[Father]

I see that I did not find instructions how to download leelaqueen odds in discord.

searching in google lead me to the following link and I am not sure if it is the LeelaQueenOdds or some old version

https://github.com/notune/LeelaQueenOdds

I see release date is 13.11.2024

Yes, that is the net used by LeelaQueenOdds from last November up until one day ago. We upgraded (?) to a new net also trained for queen odds, by "@Ira_Jed and @Lucario6607. It's not yet clear that it is stronger, but I think on balance the evidence suggests that it is a bit stronger against both human-like bots and against actual humans. Right after the switch, it played four blitz (3'0") games against GM Jaan Ehlvest, who was World number 5 about 35 years ago and is still, at age 63, rated well over the GM threshold at 2532 FIDE. Leela won three and lost one. No games decided by time, but Ehlvest did move a bit faster than necessary for a 3' game, more as if they were 2' games. He said that he thinks this was the main reason for his loss, but even so winning a blitz match at queen odds from a former top-five player indicates that the new net is not bad! By a curious coincidence, Ehlvest is the strongest player I ever defeated in a tournament game, around 1996 I believe, in the World Open.

Good evening, Mr. Larry Kaufman, gentlemen of the Leela Bots team, and Mr. Lichess. I hope you are all well and continue to do well. I wish the best for all of you and for each and every one of you. I also hope that these notes and words reach you, as well as chess enthusiasts and players, and especially those who love challenges against chess computers. Today I want to propose an idea; it is this: "Design some type of monthly open tournament or competition for each of the bots, where humans have the opportunity to play only against the bots and obtain a spot in the competition. This will increase the popularity of the bots and the incentives of the humans; it would also be reflected in the Elo within the bots' table."Sincerely, Pablo

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[pgn][/pgn][/pgn]

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[/pgn]

Good evening, Mr. Larry Kaufman. Time flies. I'm approaching my first year playing against robots, barely a month away. In this game I'm publishing, I don't understand the robot's 17th move from bishop to king's knight. I'm already facing all four robots; I'll leave the advantage knight for last. I love playing against these wonderful machines.

Good morning, Mr. Larry Kaufman and Mr. Uri Blass. Good morning, systems engineers and programmers of the LeelaBottsOdds team, and other workers and collaborators of the LeelaOdds project and participants in this forum. Good morning to all enthusiastic chess players around the world. Lichess, I write to you all only within the boundaries defined by the board itself and by the power of the pieces themselves. I hope you are all well and continue to be well. I wish each of you fair winds, fair seas, a safe voyage, and a suitcase full of hope, motivation, endeavors, and new dreams and goals in this testing life in which we head together toward infinity. I want to share another game with you that I would very much like you to take the time to study. I have sometimes considered that extremes merge into a single being: mathematics and philosophy, madness and sanity, night and day beneath the lines of dawn and dusk. Black's ninth move, King's Bishop to C5, left me perplexed. Of course, with a queen advantage in my favor, and then a bishop sacrifice, I quickly decided that my duty was not to draw but instead to beat the robot's queen-leave-odds. And so it was. I dedicate this unique game to all of you. Happy day, happy night, happy week.

[pgn][Event "casual bullet game"]
[Site "https://lichess.org/jDMVaXOE"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "jDMVaXOE"]
[UTCDate "2025.08.14"]
[UTCTime "13:48:56"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 f5 2. f4 b6 3. e3 e6 4. Nf3 Bb7 5. c3 Nf6 6. Bd3 h6 7. Qe2 Rg8 8. O-O Nc6 9. g3 Bc5 10. dxc5 g5 11. cxb6 axb6 12. Rf2 Ng4 13. Rg2 O-O-O 14. Ba6 Bxa6 15. Qxa6+ Kb8 16. Qe2 Rde8 17. Nbd2 e5 18. Nxe5 Ncxe5 19. fxe5 Nxe5 20. Nf3 Nf7 21. a4 g4 22. Nh4 Ng5 23. a5 Re6 24. axb6 Rxb6 25. Qc2 Kb7 26. Ra3 Re8 27. Rb3 Nh3+ 28. Kf1 Re6 29. Rxb6+ Rxb6 30. Rd2 Ng5 31. Rd4 Ne6 32. Rb4 Rxb4 33. cxb4 Ng5 34. Qxf5 Ne6 35. Qxg4 h5 36. Qxh5 c6 37. Qg4 Kc7 38. Qxe6 dxe6 39. Bd2 Kd6 40. Bc3 Kd5 41. Kf2 e5 42. Kf3 e4+ 43. Kf4 Kc4 44. g4 Kd5 45. g5 Ke6 46. g6 Kd5 47. g7 Kc4 48. g8=Q+ Kb5 49. Qb3 Kb6 50. Kxe4 Kc7 51. Kd4 Kb6 52. Kc4 c5 53. bxc5+ Kc6 54. Qb6+ Kd7 55. Qd6+ Kc8 56. Qe7 Kb8 57. Nf5 Ka8 58. h4 Kb8 59. h5 Ka8 60. h6 Kb8 61. h7 Ka8 62. h8=Q# { White wins by checkmate. } 1-0[/pgn]

[lkaufman]

I see that I did not find instructions how to download leelaqueen odds in discord.

searching in google lead me to the following link and I am not sure if it is the LeelaQueenOdds or some old version

https://github.com/notune/LeelaQueenOdds

I see release date is 13.11.2024

Yes, that is the net used by LeelaQueenOdds from last November up until one day ago. We upgraded (?) to a new net also trained for queen odds, by "@Ira_Jed and @Lucario6607. It's not yet clear that it is stronger, but I think on balance the evidence suggests that it is a bit stronger against both human-like bots and against actual humans. Right after the switch, it played four blitz (3'0") games against GM Jaan Ehlvest, who was World number 5 about 35 years ago and is still, at age 63, rated well over the GM threshold at 2532 FIDE. Leela won three and lost one. No games decided by time, but Ehlvest did move a bit faster than necessary for a 3' game, more as if they were 2' games. He said that he thinks this was the main reason for his loss, but even so winning a blitz match at queen odds from a former top-five player indicates that the new net is not bad! By a curious coincidence, Ehlvest is the strongest player I ever defeated in a tournament game, around 1996 I believe, in the World Open.

Good evening, Mr. Larry Kaufman, gentlemen of the Leela Bots team, and Mr. Lichess. I hope you are all well and continue to do well. I wish the best for all of you and for each and every one of you. I also hope that these notes and words reach you, as well as chess enthusiasts and players, and especially those who love challenges against chess computers. Today I want to propose an idea; it is this: "Design some type of monthly open tournament or competition for each of the bots, where humans have the opportunity to play only against the bots and obtain a spot in the competition. This will increase the popularity of the bots and the incentives of the humans; it would also be reflected in the Elo within the bots' table."Sincerely, Pablo

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[pgn][/pgn][/pgn]

[pgn]Event "casual bullet game"]
[Site "https://lichess.org/gfQneK8p"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "gfQneK8p"]
[UTCDate "2025.08.14"]
[UTCTime "03:55:33"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 b6 2. f4 Bb7 3. e3 e6 4. Nf3 h6 5. c3 g5 6. Bd3 g4 7. Nfd2 Bxg2 8. Rg1 Bb7 9. Qxg4 Nf6 10. Qe2 Nc6 11. Nf3 h5 12. Nbd2 Ng8 13. Nf1 f5 14. Bd2 Nf6 15. Ba6 Bxa6 16. Qxa6 Rb8 17. Qe2 Bg7 18. Rxg7 Kf8 19. Rg2 h4 20. h3 Ne7 21. Kf2 Ne4+ 22. Kg1 Nd5 23. Be1 Ke7 24. N1d2 Ng3 25. Bxg3 hxg3 26. Rxg3 Rbg8 27. Rxg8 Rxg8+ 28. Kh2 Nf6 29. Rg1 Rh8 30. Rg7+ Kf8 31. Qg2 Ng4+ 32. Rxg4 fxg4 33. Qxg4 Rg8 34. Qxg8+ Kxg8 35. Kg3 Kg7 36. Kg4 Kf6 37. h4 Kg6 38. h5+ Kh6 39. Nh4 a5 40. Ndf3 a4 41. a3 c5 42. Ne5 cxd4 43. exd4 d6 44. Nd7 b5 45. Nb6 Kg7 46. Kg5 Kf7 47. h6 Kg8 48. Ng6 b4 49. axb4 a3 50. bxa3 d5 51. h7+ Kxh7 52. Kf6 Kh6 53. Kxe6 Kxg6 54. Kxd5 Kf5 55. a4 Kxf4 56. a5 Ke3 57. c4 Kd3 58. c5 Ke3 59. c6 Kd3 60. c7 Ke3 61. c8=Q Kd3 62. Qc4+ Ke3 63. a6 Kf4 64. a7 Ke3 65. a8=Q Kf4 66. Qf8+ Kg3 67. Qd3+ Kh4 68. Qh7+ Kg3 69. Qhg7+ Kh4 70. Qfh8# { White wins by checkmate. } 1-0[/pgn]

Good evening, Mr. Larry Kaufman. Time flies. I'm approaching my first year playing against robots, barely a month away. In this game I'm publishing, I don't understand the robot's 17th move from bishop to king's knight. I'm already facing all four robots; I'll leave the advantage knight for last. I love playing against these wonderful machines.

Good morning, Mr. Larry Kaufman and Mr. Uri Blass. Good morning, systems engineers and programmers of the LeelaBottsOdds team, and other workers and collaborators of the LeelaOdds project and participants in this forum. Good morning to all enthusiastic chess players around the world. Lichess, I write to you all only within the boundaries defined by the board itself and by the power of the pieces themselves. I hope you are all well and continue to be well. I wish each of you fair winds, fair seas, a safe voyage, and a suitcase full of hope, motivation, endeavors, and new dreams and goals in this testing life in which we head together toward infinity. I want to share another game with you that I would very much like you to take the time to study. I have sometimes considered that extremes merge into a single being: mathematics and philosophy, madness and sanity, night and day beneath the lines of dawn and dusk. Black's ninth move, King's Bishop to C5, left me perplexed. Of course, with a queen advantage in my favor, and then a bishop sacrifice, I quickly decided that my duty was not to draw but instead to beat the robot's queen-leave-odds. And so it was. I dedicate this unique game to all of you. Happy day, happy night, happy week.

[pgn][Event "casual bullet game"]
[Site "https://lichess.org/jDMVaXOE"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "jDMVaXOE"]
[UTCDate "2025.08.14"]
[UTCTime "13:48:56"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 f5 2. f4 b6 3. e3 e6 4. Nf3 Bb7 5. c3 Nf6 6. Bd3 h6 7. Qe2 Rg8 8. O-O Nc6 9. g3 Bc5 10. dxc5 g5 11. cxb6 axb6 12. Rf2 Ng4 13. Rg2 O-O-O 14. Ba6 Bxa6 15. Qxa6+ Kb8 16. Qe2 Rde8 17. Nbd2 e5 18. Nxe5 Ncxe5 19. fxe5 Nxe5 20. Nf3 Nf7 21. a4 g4 22. Nh4 Ng5 23. a5 Re6 24. axb6 Rxb6 25. Qc2 Kb7 26. Ra3 Re8 27. Rb3 Nh3+ 28. Kf1 Re6 29. Rxb6+ Rxb6 30. Rd2 Ng5 31. Rd4 Ne6 32. Rb4 Rxb4 33. cxb4 Ng5 34. Qxf5 Ne6 35. Qxg4 h5 36. Qxh5 c6 37. Qg4 Kc7 38. Qxe6 dxe6 39. Bd2 Kd6 40. Bc3 Kd5 41. Kf2 e5 42. Kf3 e4+ 43. Kf4 Kc4 44. g4 Kd5 45. g5 Ke6 46. g6 Kd5 47. g7 Kc4 48. g8=Q+ Kb5 49. Qb3 Kb6 50. Kxe4 Kc7 51. Kd4 Kb6 52. Kc4 c5 53. bxc5+ Kc6 54. Qb6+ Kd7 55. Qd6+ Kc8 56. Qe7 Kb8 57. Nf5 Ka8 58. h4 Kb8 59. h5 Ka8 60. h6 Kb8 61. h7 Ka8 62. h8=Q# { White wins by checkmate. } 1-0[/pgn]

We have observed such inexplicable blunders by the bots on rare occasions. We are not yet sure if it is due to some bug in search contempt, some other bug, or simply a flaw in the specific network being used. We decide on algorithms and networks based on what scores the best in testing, but this won't detect a flaw that occurs in perhaps one game in a thousand. When we see such blunders we try to fix them, but usually they are not reproducible and hence impossible to fix. If it is a network-specific problem, it will disappear with the next network switch, which is frequent for queen odds but infrequent for smaller odds.

[Father]

[/quote]

Good evening, Mr. Larry Kaufman. Time flies. I'm approaching my first year playing against robots, barely a month away. In this game I'm publishing, I don't understand the robot's 17th move from bishop to king's knight. I'm already facing all four robots; I'll leave the advantage knight for last. I love playing against these wonderful machines.
[/quote]

Good morning, Mr. Larry Kaufman and Mr. Uri Blass. Good morning, systems engineers and programmers of the LeelaBottsOdds team, and other workers and collaborators of the LeelaOdds project and participants in this forum. Good morning to all enthusiastic chess players around the world. Lichess, I write to you all only within the boundaries defined by the board itself and by the power of the pieces themselves. I hope you are all well and continue to be well. I wish each of you fair winds, fair seas, a safe voyage, and a suitcase full of hope, motivation, endeavors, and new dreams and goals in this testing life in which we head together toward infinity. I want to share another game with you that I would very much like you to take the time to study. I have sometimes considered that extremes merge into a single being: mathematics and philosophy, madness and sanity, night and day beneath the lines of dawn and dusk. Black's ninth move, King's Bishop to C5, left me perplexed. Of course, with a queen advantage in my favor, and then a bishop sacrifice, I quickly decided that my duty was not to draw but instead to beat the robot's queen-leave-odds. And so it was. I dedicate this unique game to all of you. Happy day, happy night, happy week.

[pgn][Event "casual bullet game"]
[Site "https://lichess.org/jDMVaXOE"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "jDMVaXOE"]
[UTCDate "2025.08.14"]
[UTCTime "13:48:56"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 f5 2. f4 b6 3. e3 e6 4. Nf3 Bb7 5. c3 Nf6 6. Bd3 h6 7. Qe2 Rg8 8. O-O Nc6 9. g3 Bc5 10. dxc5 g5 11. cxb6 axb6 12. Rf2 Ng4 13. Rg2 O-O-O 14. Ba6 Bxa6 15. Qxa6+ Kb8 16. Qe2 Rde8 17. Nbd2 e5 18. Nxe5 Ncxe5 19. fxe5 Nxe5 20. Nf3 Nf7 21. a4 g4 22. Nh4 Ng5 23. a5 Re6 24. axb6 Rxb6 25. Qc2 Kb7 26. Ra3 Re8 27. Rb3 Nh3+ 28. Kf1 Re6 29. Rxb6+ Rxb6 30. Rd2 Ng5 31. Rd4 Ne6 32. Rb4 Rxb4 33. cxb4 Ng5 34. Qxf5 Ne6 35. Qxg4 h5 36. Qxh5 c6 37. Qg4 Kc7 38. Qxe6 dxe6 39. Bd2 Kd6 40. Bc3 Kd5 41. Kf2 e5 42. Kf3 e4+ 43. Kf4 Kc4 44. g4 Kd5 45. g5 Ke6 46. g6 Kd5 47. g7 Kc4 48. g8=Q+ Kb5 49. Qb3 Kb6 50. Kxe4 Kc7 51. Kd4 Kb6 52. Kc4 c5 53. bxc5+ Kc6 54. Qb6+ Kd7 55. Qd6+ Kc8 56. Qe7 Kb8 57. Nf5 Ka8 58. h4 Kb8 59. h5 Ka8 60. h6 Kb8 61. h7 Ka8 62. h8=Q# { White wins by checkmate. } 1-0[/pgn]
[/quote]

We have observed such inexplicable blunders by the bots on rare occasions. We are not yet sure if it is due to some bug in search contempt, some other bug, or simply a flaw in the specific network being used. We decide on algorithms and networks based on what scores the best in testing, but this won't detect a flaw that occurs in perhaps one game in a thousand. When we see such blunders we try to fix them, but usually they are not reproducible and hence impossible to fix. If it is a network-specific problem, it will disappear with the next network switch, which is frequent for queen odds but infrequent for smaller odds.
[/quote]

Good evening, Mr. Larry Kaufman.
I hope you are well and continue to do well. I also wish the best for the entire Leela team. I want to let you know that I have continued interacting with Leela's four Odds chess robots, playing a significant number of games, usually bullet games. These are: Queen, QueenForKnight, Rook, and Knight, Odds. I have observed through my experience playing against these machines that each robot has its own characteristics, although there is something in common between them, in one way or another. I would believe they share a common DNA, although a distinctly different personality. If I were asked to give a simple idea of how to define each robot, I would do so as follows:
1. LeelaQueenOdds: Blinding, roving light.
2. LeelaQueenForKnight: Constant beam of light.
3. LeelaRookOdds: Midday light.
4. LeelaKnight: Strong winds make it difficult for humans to remain standing, but even so, if humans learn to recognize the wind's behavior, they will be able to better control the fight.
I, too, have gradually learned, through experience and strength, the use of new "strategic tricks" of greater, medium, and lesser complexity that are completely new to me, and which I have been designing as I analyze the algorithms. For example, if you simplify the machine's knights in some of the bots, even though the human concedes its own diagonals as a result of the weakness of the open diagonals conceded to the computer by the machine's bishops, this in turn reduces the robot's strength and effectiveness. Despite the human's seemingly more fragile position, it actually makes it more difficult for the computer to do its job.
I also notice that LeelaQueenOdds seems to be even more popular. I would believe that LeelaKnight is a "troglodyte" who already has greater annihilating power than us modern humans. LeelaQueenOdds' personality is so prophylactic that, without knowing the mathematical reasoning, the truth is that she is more effective than the robots Rook and Queen for Knight. I don't know the explanation, but that machine is like a wasp that you think is dead and then stings you with its stinger and injects you with poison. I have also seen the characteristics of the humans who have faced this robot, and I observe that the level of competition is higher on the QueenOdds leaderboard; there, it seems to me, there is a broader list of professional federated chess players. Today, GM Joel Benjamin, "SuperEasy," retook the top spot on the LeelaQueenOdds leaderboard, which makes me very happy. I currently hold first place in the leaderboard for the Rook and QueenForKnight robots, and 20th place for LeelaQueenOdds and 19th place for LeelaKnightOdds. I think I can still improve relatively significantly in Queen and Knight Odds, but little by little. What I don't know is if I can maintain my first place in LeelaRook and LeelaQueenForKnight in the future. For now, I see that at bullet level, I am the highest-ranked human in the Knight, Rook, and QueenForKnight robots, but I don't know if this will be sustainable. I've been analyzing the reason for these positions in bullet, and I think I have the answer that explains it. It is this: In the case of the LeelaQueenOdds robot, the handicap is so high that a high-level human player means that the human's skill prevails in the balance of combat even under orthodox systems. However, in the cases of rook, queen for knight, and even knight, the unorthodox system I proposed along strategic lines prevails over the orthodox systems of classical mathematical coherence paradigms. But here a very interesting challenge arises for programmers: to make the QueenOdds robot more and more efficient, to the point that it can overcome, in very fast games, the conventional game paradigms applied by human chess stars. I don't want to go on any longer, Dr. Larry Kaufman. I'm becoming tiresome and tedious, but I hope I have been of help to all of you in the work and growth of robots. Likewise, I look forward to departing into eternity on the train and station that the programmer of the universe assigns to me—obviously as a surprise, since neither the day nor the hour of the journey is revealed to us. I hope to continue battling against all these chess machines and to be a challenge for all of you to overcome through your beloved chess machines. The fuel that feeds my soul is green; its name is "Hope." Happy night and day to all of you. Thank you for the machines!

[Father]

Good evening, Mr. Larry Kaufman. Time flies. I'm approaching my first year playing against robots, barely a month away. In this game I'm publishing, I don't understand the robot's 17th move from bishop to king's knight. I'm already facing all four robots; I'll leave the advantage knight for last. I love playing against these wonderful machines.
[/quote]

Good morning, Mr. Larry Kaufman and Mr. Uri Blass. Good morning, systems engineers and programmers of the LeelaBottsOdds team, and other workers and collaborators of the LeelaOdds project and participants in this forum. Good morning to all enthusiastic chess players around the world. Lichess, I write to you all only within the boundaries defined by the board itself and by the power of the pieces themselves. I hope you are all well and continue to be well. I wish each of you fair winds, fair seas, a safe voyage, and a suitcase full of hope, motivation, endeavors, and new dreams and goals in this testing life in which we head together toward infinity. I want to share another game with you that I would very much like you to take the time to study. I have sometimes considered that extremes merge into a single being: mathematics and philosophy, madness and sanity, night and day beneath the lines of dawn and dusk. Black's ninth move, King's Bishop to C5, left me perplexed. Of course, with a queen advantage in my favor, and then a bishop sacrifice, I quickly decided that my duty was not to draw but instead to beat the robot's queen-leave-odds. And so it was. I dedicate this unique game to all of you. Happy day, happy night, happy week.

[pgn][Event "casual bullet game"]
[Site "https://lichess.org/jDMVaXOE"]
[Date "2025.08.14"]
[White "Catecan"]
[Black "LeelaQueenOdds"]
[Result "1-0"]
[GameId "jDMVaXOE"]
[UTCDate "2025.08.14"]
[UTCTime "13:48:56"]
[WhiteElo "2022"]
[BlackElo "2000"]
[BlackTitle "BOT"]
[Variant "From Position"]
[TimeControl "60+0"]
[ECO "?"]
[Opening "?"]
[Termination "Normal"]
[FEN "rnb1kbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
[SetUp "1"]
[Annotator "lichess.org"]

1. d4 f5 2. f4 b6 3. e3 e6 4. Nf3 Bb7 5. c3 Nf6 6. Bd3 h6 7. Qe2 Rg8 8. O-O Nc6 9. g3 Bc5 10. dxc5 g5 11. cxb6 axb6 12. Rf2 Ng4 13. Rg2 O-O-O 14. Ba6 Bxa6 15. Qxa6+ Kb8 16. Qe2 Rde8 17. Nbd2 e5 18. Nxe5 Ncxe5 19. fxe5 Nxe5 20. Nf3 Nf7 21. a4 g4 22. Nh4 Ng5 23. a5 Re6 24. axb6 Rxb6 25. Qc2 Kb7 26. Ra3 Re8 27. Rb3 Nh3+ 28. Kf1 Re6 29. Rxb6+ Rxb6 30. Rd2 Ng5 31. Rd4 Ne6 32. Rb4 Rxb4 33. cxb4 Ng5 34. Qxf5 Ne6 35. Qxg4 h5 36. Qxh5 c6 37. Qg4 Kc7 38. Qxe6 dxe6 39. Bd2 Kd6 40. Bc3 Kd5 41. Kf2 e5 42. Kf3 e4+ 43. Kf4 Kc4 44. g4 Kd5 45. g5 Ke6 46. g6 Kd5 47. g7 Kc4 48. g8=Q+ Kb5 49. Qb3 Kb6 50. Kxe4 Kc7 51. Kd4 Kb6 52. Kc4 c5 53. bxc5+ Kc6 54. Qb6+ Kd7 55. Qd6+ Kc8 56. Qe7 Kb8 57. Nf5 Ka8 58. h4 Kb8 59. h5 Ka8 60. h6 Kb8 61. h7 Ka8 62. h8=Q# { White wins by checkmate. } 1-0[/pgn]
[/quote]

We have observed such inexplicable blunders by the bots on rare occasions. We are not yet sure if it is due to some bug in search contempt, some other bug, or simply a flaw in the specific network being used. We decide on algorithms and networks based on what scores the best in testing, but this won't detect a flaw that occurs in perhaps one game in a thousand. When we see such blunders we try to fix them, but usually they are not reproducible and hence impossible to fix. If it is a network-specific problem, it will disappear with the next network switch, which is frequent for queen odds but infrequent for smaller odds.
[/quote]

Good evening, Mr. Larry Kaufman.
I hope you are well and continue to do well. I also wish the best for the entire Leela team. I want to let you know that I have continued interacting with Leela's four Odds chess robots, playing a significant number of games, usually bullet games. These are: Queen, QueenForKnight, Rook, and Knight, Odds. I have observed through my experience playing against these machines that each robot has its own characteristics, although there is something in common between them, in one way or another. I would believe they share a common DNA, although a distinctly different personality. If I were asked to give a simple idea of how to define each robot, I would do so as follows:
1. LeelaQueenOdds: Blinding, roving light.
2. LeelaQueenForKnight: Constant beam of light.
3. LeelaRookOdds: Midday light.
4. LeelaKnight: Strong winds make it difficult for humans to remain standing, but even so, if humans learn to recognize the wind's behavior, they will be able to better control the fight.
I, too, have gradually learned, through experience and strength, the use of new "strategic tricks" of greater, medium, and lesser complexity that are completely new to me, and which I have been designing as I analyze the algorithms. For example, if you simplify the machine's knights in some of the bots, even though the human concedes its own diagonals as a result of the weakness of the open diagonals conceded to the computer by the machine's bishops, this in turn reduces the robot's strength and effectiveness. Despite the human's seemingly more fragile position, it actually makes it more difficult for the computer to do its job.
I also notice that LeelaQueenOdds seems to be even more popular. I would believe that LeelaKnight is a "troglodyte" who already has greater annihilating power than us modern humans. LeelaQueenOdds' personality is so prophylactic that, without knowing the mathematical reasoning, the truth is that she is more effective than the robots Rook and Queen for Knight. I don't know the explanation, but that machine is like a wasp that you think is dead and then stings you with its stinger and injects you with poison. I have also seen the characteristics of the humans who have faced this robot, and I observe that the level of competition is higher on the QueenOdds leaderboard; there, it seems to me, there is a broader list of professional federated chess players. Today, GM Joel Benjamin, "SuperEasy," retook the top spot on the LeelaQueenOdds leaderboard, which makes me very happy. I currently hold first place in the leaderboard for the Rook and QueenForKnight robots, and 20th place for LeelaQueenOdds and 19th place for LeelaKnightOdds. I think I can still improve relatively significantly in Queen and Knight Odds, but little by little. What I don't know is if I can maintain my first place in LeelaRook and LeelaQueenForKnight in the future. For now, I see that at bullet level, I am the highest-ranked human in the Knight, Rook, and QueenForKnight robots, but I don't know if this will be sustainable. I've been analyzing the reason for these positions in bullet, and I think I have the answer that explains it. It is this: In the case of the LeelaQueenOdds robot, the handicap is so high that a high-level human player means that the human's skill prevails in the balance of combat even under orthodox systems. However, in the cases of rook, queen for knight, and even knight, the unorthodox system I proposed along strategic lines prevails over the orthodox systems of classical mathematical coherence paradigms. But here a very interesting challenge arises for programmers: to make the QueenOdds robot more and more efficient, to the point that it can overcome, in very fast games, the conventional game paradigms applied by human chess stars. I don't want to go on any longer, Dr. Larry Kaufman. I'm becoming tiresome and tedious, but I hope I have been of help to all of you in the work and growth of robots. Likewise, I look forward to departing into eternity on the train and station that the programmer of the universe assigns to me—obviously as a surprise, since neither the day nor the hour of the journey is revealed to us. I hope to continue battling against all these chess machines and to be a challenge for all of you to overcome through your beloved chess machines. The fuel that feeds my soul is green; its name is "Hope." Happy night and day to all of you. Thank you for the machines!
[/quote]

Good evening, Mr. Larry Kaufman. The word to define today's chess tournament, between my imaginary wooden Knight "Catecan" and me against the robot LeelaKnight, can be summed up in one word: "Exhausted." We just need to improve our rankings on the LeelaKnightOdds and LeelaQueenOdds tables. Here's how it goes:
1. LeelaQueenOdds: Catecan is ranked nineteenth (19th) at the time of writing.
2. LeelaKnightForKnight: Catecan is ranked first (1st) at the time of writing.
3. LeelaRookOdds: Catecan is ranked first (1st) at the time of writing.
4. LeelaKnight: Catecan is ranked ninth (9th) at the time of writing.
I think we'll have to rest both rider and mount. The effort has been great, and the LeelaKnight robot has hit my self-esteem really hard. Then, very calmly, I'll take the LeelaQueenOdds robot and probably move up the leaderboard, although of course I don't have the strength to be number one there.
The only way I can think of for now for me to aspire to occupy top spots on the LeelaQueenOdds robot leaderboard would be for the robot to acquire new strengths that make it much more lethal. I'd like to ask something: Can LeelaQueenOdds substantially increase its playing strength against the best in the bullet leaderboard to the point of hitting them really hard and strong? If so, I believe it would benefit me greatly. Thanks again.

Así estan las tablas hoy:

1. LeelaknightOdds:

Leela Bots Leaderboard
Elo Ratings and Statistics
LeelaQueenOdds
LeelaQueenForKnight
LeelaRookOdds
LeelaKnightOdds
Rank Player Elo W-D-L Median Time Control Last Game
1 limpopo777 2522 38-115-186 20+3 2025-08-12
2 KarlBallmer 2403 15-113-390 3+2 2025-06-09
3 drop_stone 2354 21-119-1094 1+1 2025-08-19
4 Former_Player 2262 17-138-742 10+3 2025-07-04
5 qwerty1234chess 2224 10-3-7 8+2 2025-08-08
6 kingside_master 2110 16-0-33 30+30 2025-07-17
7 a2c4 2088 12-1-13 15+10 2025-05-08
8 FeegLood 2069 7-7-33 3+1 2025-08-05
9 Catecan 2051 1-73-701 1+0 2025-08-20
10 likigor 2050 11-3-145 5+3 2025-04-04
11 Baba_Bulkin 2048 8-2-35 25+20 2025-08-17
12 Aczgego 2037 0-80-389 5+5 2025-07-06
13 ENajer 2028 1-38-94 5+3 2025-05-26
14 Themathic 2005 5-20-600 3+2 2025-08-07
15 N0um 2000 5-0-3 5+3 2025-07-10
16 trenersahnutyi 1998 5-0-27 10+5 2025-05-18
17 testxxx478r 1990 9-3-6 10+0 2025-06-03
18 CorvoRosso 1984 16-7-254 3+6 2025-08-11
19 EvilGenius94 1969 4-1-18 5+3 2025-08-01
20 Lione38 1960 2-45-1208 5+3 2025-07-06


2- LeelaQueenOdds:

Leela Bots Leaderboard
Elo Ratings and Statistics
LeelaQueenOdds
LeelaQueenForKnight
LeelaRookOdds
LeelaKnightOdds
Rank Player Elo W-D-L Median Time Control Last Game
1 SuperEasy 2968 506-6-16 8+3 2025-08-19
2 Arka50 2966 85-27-66 1+0 2025-06-20
3 FeegLood 2949 170-65-163 1+0 2025-07-26
4 wateenellende 2870 102-72-173 1+0 2025-07-17
5 InvisibleGuest2023 2823 107-127-160 1+0 2025-06-09
6 blindcrocodile 2822 156-25-170 1+1 2025-08-19
7 JoanFluvia 2808 443-58-329 3+1 2025-08-14
8 Former_Player 2773 245-44-119 3+2 2025-06-25
9 verystablegenius69 2768 81-73-230 1+0 2025-05-07
10 VladimirMarkovic 2733 405-292-1088 1+0 2025-06-25
11 AdmiralHabibi 2727 77-7-65 3+2 2025-07-10
12 edgelorde 2723 898-190-724 1+1 2025-06-16
13 dauANHbac 2702 46-1-32 1+1 2025-08-08
14 jflup 2683 137-14-117 3+1 2025-08-05
15 Alex_Muezersky 2672 59-27-93 1+1 2025-04-05
16 Orichess88 2662 54-2-33 2+3 2025-07-25
17 zizita123 2662 68-16-57 2+1 2025-07-24
18 Phara0h1 2654 479-88-463 5+0 2025-06-11
19 Catecan 2642 13-5617-4696 1+0 2025-08-20
20 Dobryfen 2642 70-5-43 3+2 2025-08-17


3- LeelaRookOdds:

Leela Bots Leaderboard
Elo Ratings and Statistics
LeelaQueenOdds
LeelaQueenForKnight
LeelaRookOdds
LeelaKnightOdds
Rank Player Elo W-D-L Median Time Control Last Game
1 Catecan 2619 0-280-526 1+0 2025-07-25
2 Iron_jack 2608 68-9-187 10+5 2025-05-19
3 JungleHook 2544 63-98-1376 5+2 2025-08-13
4 StepanOsinovsky 2481 46-17-174 11+13 2025-08-08
5 drop_stone 2416 26-56-296 1+1 2025-07-15
6 FeegLood 2295 12-5-22 3+2 2025-07-29
7 Omnivor 2281 21-29-174 5+5 2025-07-26
8 Leeladestroyer2 2139 15-4-27 30+30 2025-07-10
9 UgoFantozzi 2112 16-7-281 5+5 2025-07-14
10 OhNoBaby 2105 26-36-285 5+2 2025-07-03
11 Somelier 2095 20-73-416 3+5 2025-04-04
12 likigor 2047 6-1-26 5+3 2025-04-04
13 CoolerMove 2046 9-12-107 10+11 2025-07-22
14 lanturn17 2039 6-0-6 5+3 2025-06-01
15 Capadjo 2038 4-12-25 5+3 2025-04-18
16 Gector1Discharge 2025 6-2-12 30+10 2025-08-07
17 Caissaan 2015 14-12-265 3+2 2025-08-02
18 Andrejs21 2013 7-5-43 3+2 2025-06-07
19 WikiChess 2010 8-2-79 5+3 2025-08-19
20 MrMasabo 2002 9-5-175 3+2 2025-08-04

4- LeelaQueenForKnight:

Leela Bots Leaderboard
Elo Ratings and Statistics
LeelaQueenOdds
LeelaQueenForKnight
LeelaRookOdds
LeelaKnightOdds
Rank Player Elo W-D-L Median Time Control Last Game
1 Catecan 2529 1-232-247 1+0 2025-08-17
2 ygj 2500 36-11-104 20+3 2025-08-19
3 slowpoke28 2079 6-4-0 5+3 2025-07-11
4 afish 2028 5-10-39 3+3 2025-03-13
5 wannabe2700 2009 5-1-7 3+2 2025-08-12
6 parfumer1 1993 6-2-30 3+2 2025-08-19
7 Alpha_Zero_System 1982 6-8-56 5+3 2025-08-03
8 Chess-Network 1969 4-1-9 3+2 2025-06-01
9 SIDDHARTH_CHESS 1945 2-7-24 1+0 2025-07-31
10 capitansalami01 1934 5-0-43 1+4 2025-06-05
11 Sir_Patzer 1919 3-0-12 3+3 2025-08-05
12 Hissha 1899 2-2-8 1+10 2025-07-22
13 TotientQuotient 1897 5-10-698 5+3 2025-06-13
14 Mazacrola 1888 2-1-12 9+1 2025-07-09
15 RetiredProffesor 1888 1-5-11 5+5 2025-07-23
16 meta_ing112 1881 2-1-35 10+5 2025-07-01
17 AstanehChess 1880 2-0-2 5+3 2025-03-29
18 d4Nf6Bg51-0 1880 2-1-39 15+15 2025-05-07
19 Permutohedron 1880 2-0-0 1+5 2025-08-06
20 John1914 1879 2-0-25 3+2 2025-07-03

[lkaufman]

The only way I can think of for now for me to aspire to occupy top spots on the LeelaQueenOdds robot leaderboard would be for the robot to acquire new strengths that make it much more lethal. I'd like to ask something: Can LeelaQueenOdds substantially increase its playing strength against the best in the bullet leaderboard to the point of hitting them really hard and strong? If so, I believe it would benefit me greatly. Thanks again.]

LeelaQueenOdds was upgraded late night Aug. 18 to a new neural net, trained by @Ira_Jed, based on 350,000 queen odds games against human-simulating bots, much more than ever before used. My testing showed it to perform 73 elo better giving queen odds than the net previously used. Results vs humans do already seem better. We'll have to see if the improvement is enough to make a difference against the superstar players who top the Leaderboard. LeelaPieceOdds was upgraded this morning to use this same net for all the odds of queen plus anything, as testing showed that it improved over the previously used net by a substantial margin in most cases. It was also upgraded to use the same new search (by Lucario6607) as the other bots are now using, which boosted them all by forty to fifty elo. I was hoping that the new net would also be better at queen for knight odds, but testing did not prove that to be so, so no change to that bot.

[Father]

The only way I can think of for now for me to aspire to occupy top spots on the LeelaQueenOdds robot leaderboard would be for the robot to acquire new strengths that make it much more lethal. I'd like to ask something: Can LeelaQueenOdds substantially increase its playing strength against the best in the bullet leaderboard to the point of hitting them really hard and strong? If so, I believe it would benefit me greatly. Thanks again.]

LeelaQueenOdds was upgraded late night Aug. 18 to a new neural net, trained by @Ira_Jed, based on 350,000 queen odds games against human-simulating bots, much more than ever before used. My testing showed it to perform 73 elo better giving queen odds than the net previously used. Results vs humans do already seem better. We'll have to see if the improvement is enough to make a difference against the superstar players who top the Leaderboard. LeelaPieceOdds was upgraded this morning to use this same net for all the odds of queen plus anything, as testing showed that it improved over the previously used net by a substantial margin in most cases. It was also upgraded to use the same new search (by Lucario6607) as the other bots are now using, which boosted them all by forty to fifty elo. I was hoping that the new net would also be better at queen for knight odds, but testing did not prove that to be so, so no change to that bot.

Thank you very much Mr. Larry Kaufman for your information. I would like to ask you, given the current strength of the four robots in FIDE ratings, what would you believe the tentative FIDE rating of the four robots could be in time controls of: 3 plus 2; 3-0; 1 plus 1; 1-0; 5-0; 10-0; 20-0, and in the regulatory FIDE time used in official tournaments for the World Human Championship. Thank you Mr. Larry Kaufman. Note: Personally, I have noticed, since yesterday and today, that the robots Knight and Queen are stronger. I don't know if it's just my imagination, but that's my perception. Taking a simple draw against LeelaKnight is costing me a lot of difficulty, even though I vary my positions.

[lkaufman]

The only way I can think of for now for me to aspire to occupy top spots on the LeelaQueenOdds robot leaderboard would be for the robot to acquire new strengths that make it much more lethal. I'd like to ask something: Can LeelaQueenOdds substantially increase its playing strength against the best in the bullet leaderboard to the point of hitting them really hard and strong? If so, I believe it would benefit me greatly. Thanks again.]

LeelaQueenOdds was upgraded late night Aug. 18 to a new neural net, trained by @Ira_Jed, based on 350,000 queen odds games against human-simulating bots, much more than ever before used. My testing showed it to perform 73 elo better giving queen odds than the net previously used. Results vs humans do already seem better. We'll have to see if the improvement is enough to make a difference against the superstar players who top the Leaderboard. LeelaPieceOdds was upgraded this morning to use this same net for all the odds of queen plus anything, as testing showed that it improved over the previously used net by a substantial margin in most cases. It was also upgraded to use the same new search (by Lucario6607) as the other bots are now using, which boosted them all by forty to fifty elo. I was hoping that the new net would also be better at queen for knight odds, but testing did not prove that to be so, so no change to that bot.

Thank you very much Mr. Larry Kaufman for your information. I would like to ask you, given the current strength of the four robots in FIDE ratings, what would you believe the tentative FIDE rating of the four robots could be in time controls of: 3 plus 2; 3-0; 1 plus 1; 1-0; 5-0; 10-0; 20-0, and in the regulatory FIDE time used in official tournaments for the World Human Championship. Thank you Mr. Larry Kaufman. Note: Personally, I have noticed, since yesterday and today, that the robots Knight and Queen are stronger. I don't know if it's just my imagination, but that's my perception. Taking a simple draw against LeelaKnight is costing me a lot of difficulty, even though I vary my positions.

Well, as I said LeelaQueenOdds did receive a major upgrade late on Aug 18, so it's not surprising you see a difference. Just a few hours ago the LeelaPieceOdds bot was upgraded across the board; the new search was added, and all the multi-piece handicaps were given new nets and settings, in some cases with a massive gain in strength (mostly at the huge handicaps). Knight odds and Rook odds still use the same net as they did at the end of 2024, but they do use the new search which might be a 50 elo boost, so that is probably what you are observing at knight odds. The switch to a 5090 and to using onnx-trt back in June also boosted all the bots, settings have been refined, and the opening books are updated often. These things add up, and results just keep getting better. It's too soon to know the full effect of these recent changes. Probably LeelaQueenOdds is around FIDE 2400 now at 3'2", meaning that is the rating needed to break even. But note that doesn't mean it could get a 2400 rating against much lower rated players if they play for a draw (as you do), because Leela isn't programmed to avoid draws unless you have made a significant error, she assumes you are strong enough to score 50% at the odds. Not enough data for LeelaQueenforKnight to be very precise, probably adding 300 to the queen odds rating is about right. Leela rook odds is clearly above human level at 3'2", perhaps about 2950 FIDE now. Leela Knight odds is perhaps 3100 FIDE at 3'2" blitz if playing against 2800 rated players. But if it had to play FIDE 2000s every game, and they all played for draws, it might only get 2400 or so. Perhaps the best way to think of these estimated ratings is to imagine that all games were played with a rule that in the event of a draw, the game would be replayed until someone won, with the draws not counting (19th century rules!). As for other time limits, the adjustments I gave a few months ago should be the same. At classical time limits, we only know that we performed about 2425 FIDE at knight odds vs Joel Benjamin in January, but improvements are enough that I believe we would do better than that today even at rook odds, much better at knight, so maybe 2450 FIDE at rook, 2600 at knight. Perhaps 1900 at queen?