Does Leela Odds bots know to play in a different way in different time controls?

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Uri Blass
Posts: 10790
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Does Leela Odds bots know to play in a different way in different time controls?

Post by Uri Blass »

The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
Lunar
Posts: 12
Joined: Wed May 21, 2025 12:32 pm
Full name: Patrick Hilhorst

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by Lunar »

(Hi all, first post, etc.)

Larry Kaufman (or at least someone who claims to be him in a Youtube handle) had this to say (emphasis mine):
@Valyxn No, it doesn't currently consider time or the opponent, it just assumes that the opponent is of a specific level (depending on the odds). That level might correspond for queen odds to a human GM playing bullet or an IM playing blitz or an FM playing Rapid, for example. If it knew the remaining clock time, the increment, and the opponent's correct Elo, it might be able to play even better with enough work, but without all three it's not clear that it would help much, except of course if there isn't any increment and not enough remaining time to mate easily, it should just play to delay the mate, but it usually will do that anyway. It would gain a little by refusing all draws by repetition if there wasn't enough time for the opponent to mate in any case.
He had some very interresting comments (https://www.youtube.com/watch?v=pYO9w3t ... z8t4AaABAg) in a thread on a Youtube video where Hikaru plays against Leela with Knight odds.
Father
Posts: 1809
Joined: Sun Mar 19, 2006 4:39 am
Location: Colombia
Full name: Pablo Ignacio Restrepo

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by Father »

Lunar wrote: Thu Jun 05, 2025 7:59 pm (Hi all, first post, etc.)

Larry Kaufman (or at least someone who claims to be him in a Youtube handle) had this to say (emphasis mine):
@Valyxn No, it doesn't currently consider time or the opponent, it just assumes that the opponent is of a specific level (depending on the odds). That level might correspond for queen odds to a human GM playing bullet or an IM playing blitz or an FM playing Rapid, for example. If it knew the remaining clock time, the increment, and the opponent's correct Elo, it might be able to play even better with enough work, but without all three it's not clear that it would help much, except of course if there isn't any increment and not enough remaining time to mate easily, it should just play to delay the mate, but it usually will do that anyway. It would gain a little by refusing all draws by repetition if there wasn't enough time for the opponent to mate in any case.
He had some very interresting comments (https://www.youtube.com/watch?v=pYO9w3t ... z8t4AaABAg) in a thread on a Youtube video where Hikaru plays against Leela with Knight odds.
This information is very interesting. I think it's necessary to consider the color the robot operates under, as well as the time controls. In fact, in the leaderboard of the top 100 humans facing the LeelaQueenOdds robot, we can see a variety of time controls. It would also be interesting to calculate the weighted average of the ratings from the leaderboard of the top 100 humans compared to the probable Fide and Lichess ratings. At 2900 there are 2 players, at 2800 right now there are five human players, at 2700 we are currently in the table thirteen players, that is, 13%, above the 2600 elo of the table today there are 19%, above 2500 elo 36%, above 2400 elo there are already 53% of humans !!, that is to say the table is already very powerful in itself considered, not to mention players above 2300 elo 92% of humans, and the lowest elo of the table is 2284 elo !! So then just appearing in the table of the 100 best humans that face the robot LeelaQueenOdds is becoming increasingly difficult. I think it would be very interesting to design some kind of tournament where humans without facing each other, we only face the robot, and thus determine the position of each human in the competition. I think it should be by time controls, so from bullet tournaments to classic controls. Here's the current table showing the top 100 humans facing LeelaQueenOdds as of today:

LeelaQueenOdds Leaderboard
Next update in: 1m 37s
# Player Rating Games Last Game TC
1 Arka50 2957 122 2025-05-09 1+0
2 SuperEasy 2940 406 2025-06-04 8+3
3 verystablegenius69 2862 384 2025-05-07 1+0
4 VladimirMarkovic 2819 1387 2025-06-05 1+0
5 wateenellende 2814 154 2025-03-26 1+0
6 InvisibleGuest2023 2795 232 2025-05-26 1+0
7 Former_Player 2778 308 2025-06-01 3+2
8 JoanFluvia 2774 631 2025-06-04 2+1
9 Mikulas 2764 591 2025-06-03 2+1
10 edgelorde 2740 1615 2025-06-05 1+1
11 Hissha 2723 135 2025-05-27 1+10
12 jflup 2705 175 2025-06-04 3+1
13 Catecan 2701 6495 2025-06-05 1+0
14 Orichess88 2694 88 2025-04-03 2+3
15 Alex_Muezersky 2673 179 2025-04-05 1+1
16 dauANHbac 2669 74 2025-04-27 1+1
17 smeets 2662 29 2025-04-02 2+1
18 Phara0h1 2662 995 2025-05-28 4+0
19 FeegLood 2626 108 2025-06-04 1+0
20 Paulewski 2580 333 2025-06-05 5+0
21 pozvonochek 2573 187 2025-03-13 1+0
22 jtcoach 2563 185 2025-03-22 3+2
23 VTChess 2557 1704 2025-06-04 1+1
24 pultis12 2541 259 2025-04-02 1+0
25 gotschi 2536 1387 2025-06-03 2+3
26 Oleg_Papayan 2530 153 2025-05-11 2+0
27 FinalSpot 2527 125 2025-03-29 3+2
28 leelan1frenemy 2523 337 2025-06-05 3+1
29 arturchix 2520 38 2025-03-27 2+1
30 Andrey_Smaev_1998 2511 222 2025-04-24 7+7
31 FourDWorm 2507 50 2025-06-05 5+3
32 turton 2507 407 2025-06-05 4+7
33 BT_CHESS42 2507 429 2025-05-29 3+2
34 Iron_jack 2506 290 2025-05-20 5+2
35 monkeyking123 2505 132 2025-05-04 1+0
36 vilius100 2503 175 2025-06-05 3+2
37 eelke 2497 124 2025-04-17 1+1
38 davidb2 2475 127 2025-03-15 1+0
39 Omnivor 2472 30 2025-05-14 5+5
40 ITakeTrenPassant 2462 200 2025-03-27 3+1
41 LeafTree 2458 2961 2025-06-03 10+0
42 Dobryfen 2458 75 2025-06-05 3+2
43 MiguelPetroff 2451 165 2025-05-04 3+2
44 Fumitoks 2446 1475 2025-06-05 1+0
45 quetag 2439 170 2025-06-04 5+3
46 vlahi 2427 30 2025-03-30 3+0
47 shreckm8 2427 429 2025-04-04 3+0
48 AsheliaXII 2418 234 2025-06-01 15+10
49 kristie 2412 143 2025-03-28 1+0
50 resb 2406 103 2025-05-31 5+3
51 Mykingislonely 2405 277 2025-04-03 2+0
52 zerture 2403 331 2025-06-01 8+3
53 tasknight 2400 469 2025-04-10 2+2
54 edgebottle 2399 1584 2025-05-11 5+5
55 Spinaltap 2398 52 2025-03-23 3+0
56 garyhat 2398 135 2025-03-27 3+0
57 cryptoenthusiast 2393 71 2025-05-14 1+1
58 ChessEducation 2390 178 2025-06-03 3+2
59 Sersemlet 2373 138 2025-02-27 3+2
60 astrozilla 2373 340 2025-04-22 3+2
61 xmyy080909 2372 301 2025-04-02 1+1
62 ioste29 2371 61 2025-03-19 3+1
63 behemothbeer 2371 508 2025-05-11 5+3
64 Goranor 2368 39 2025-05-31 4+4
65 CoolerMove 2366 56 2025-06-04 10+10
66 karloballa 2366 222 2025-03-30 5+2
67 Litigator94 2361 43 2025-03-08 3+2
68 White_Phoenix 2360 91 2025-03-30 5+3
69 CatAdams 2360 701 2025-06-03 3+2
70 AdmiralHabibi 2358 102 2025-06-05 1+0
71 fryingdog 2353 2238 2025-05-27 4+5
72 klfo 2353 204 2025-06-04 4+4
73 Reuseup 2350 481 2025-06-04 5+5
74 Switchitter 2347 127 2025-04-23 3+1
75 benkonian 2344 108 2025-03-30 1+1
76 Chillkroete77 2341 126 2025-05-15 3+2
77 slowpoke28 2337 73 2025-05-11 3+2
78 darvec112357 2332 47 2025-05-10 5+3
79 Somelier 2330 30 2025-03-18 2+2
80 nauthnim 2328 471 2025-03-31 3+2
81 dritharashtrar 2328 42 2025-03-17 3+1
82 Kolibriechess 2324 77 2025-05-26 10+6
83 grandmastergauri 2323 74 2025-03-23 1+1
84 JeanLassalle64 2322 25 2025-05-31 3+2
85 Zaxar2010 2317 45 2025-04-29 5+3
86 Kelvandil 2314 84 2025-03-24 3+2
87 Chrisinc 2312 29 2025-04-14 15+10
88 SomeCM 2311 305 2025-06-02 3+0
89 FinalCut 2311 35 2025-05-27 5+0
90 Inventor_1 2309 35 2025-04-16 5+17
91 Red_Aruba 2307 145 2025-05-08 3+5
92 sergo786 2303 437 2025-04-18 5+3
93 King_Lenny 2298 393 2025-05-09 1+0
94 rowrulz 2297 23 2025-05-31 5+2
95 Srive 2293 62 2025-03-20 5+3
96 Dalnor 2291 601 2025-06-04 9+3
97 Carobee 2291 30 2025-04-10 1+2
98 JindrichPavelka 2290 246 2025-06-05 3+2
99 so197371so 2284 263 2025-04-12 3+2
100 Finnish_wall 2284 56 2025-05-14 9+2
I am thinking chess is in a coin.Human beings for ever playing in one face.Now I am playing in the other face:"Antichess". Computers are as a fortres where owner forgot to close a little door behind. You must enter across this door.Forget the front.
lkaufman
Posts: 6227
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by lkaufman »

Uri Blass wrote: Thu Jun 05, 2025 12:22 pm The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
In theory we could have multiple nets based on time control, with each net trained against a bot that was itself trained to simulate humans of a specified level at a specified time control, but the benefit would not justify the massive amount of both human and machine time needed. We are continuing to make progress in multiple ways. The White Queen odds was updated a few days ago to a new net that performs slightly better against bots at least, with a new net for Black likely in a week or so. The rook odds and knight odds were transferred from a 4090 based machine to a 5090 based machine which allowed for a 50% increase in max nodes with a similar average time per move; it seems to be performing even better with this change. The two knights handicap was switched to the new queen odds net, which somehow tested as more of an improvement for two knights than for queen (!), so I particularly recommend two knights odds (using "LeelaPieceOdds") for very strong players who find queen odds a bit too easy (at their preferred time control) but who hardly ever win at rook odds. Perhaps soon two knights can be a fair match in 3'2" blitz for even rather strong grandmasters. That is a sentence I would never have imagined myself ever writing a year ago! Interestingly, the low-rated GM who set off the boom in interest in LeelaQueenOdds when he lost his first five blitz (3'2") at queen odds, "Mikulas", eventually pulled ahead after hundreds of games, and made a nearly even score after that at 2'2", and is now number 9 on the Leaderboard!
Komodo rules!
Father
Posts: 1809
Joined: Sun Mar 19, 2006 4:39 am
Location: Colombia
Full name: Pablo Ignacio Restrepo

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by Father »

lkaufman wrote: Fri Jun 06, 2025 3:41 am
Uri Blass wrote: Thu Jun 05, 2025 12:22 pm The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
In theory we could have multiple nets based on time control, with each net trained against a bot that was itself trained to simulate humans of a specified level at a specified time control, but the benefit would not justify the massive amount of both human and machine time needed. We are continuing to make progress in multiple ways. The White Queen odds was updated a few days ago to a new net that performs slightly better against bots at least, with a new net for Black likely in a week or so. The rook odds and knight odds were transferred from a 4090 based machine to a 5090 based machine which allowed for a 50% increase in max nodes with a similar average time per move; it seems to be performing even better with this change. The two knights handicap was switched to the new queen odds net, which somehow tested as more of an improvement for two knights than for queen (!), so I particularly recommend two knights odds (using "LeelaPieceOdds") for very strong players who find queen odds a bit too easy (at their preferred time control) but who hardly ever win at rook odds. Perhaps soon two knights can be a fair match in 3'2" blitz for even rather strong grandmasters. That is a sentence I would never have imagined myself ever writing a year ago! Interestingly, the low-rated GM who set off the boom in interest in LeelaQueenOdds when he lost his first five blitz (3'2") at queen odds, "Mikulas", eventually pulled ahead after hundreds of games, and made a nearly even score after that at 2'2", and is now number 9 on the Leaderboard!
Thank you, Mr. Larry Kaufman, for your teachings. I've been analyzing the new humans who are steadily rising on the LeelaQueenOdds leaderboard, and I'm placing special emphasis on one man I find "super gifted." This is GM "FeegLood." Every day, belonging to or appearing on the leaderboard of the top 100 humans becomes more difficult and meritorious. I confess that I would be very sad if my ideal horse, "Catecan," and my own self were to be forgotten. I ask God for the strength to endure on the leaderboard. I would like to be on the same level as the giants Nakamura and Magnus; then I would be ready to travel to heaven on my horse, Catecan. But truth be told, this leaderboard describes another form of greatness, a giant in itself.
I am thinking chess is in a coin.Human beings for ever playing in one face.Now I am playing in the other face:"Antichess". Computers are as a fortres where owner forgot to close a little door behind. You must enter across this door.Forget the front.
Uri Blass
Posts: 10790
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by Uri Blass »

lkaufman wrote: Fri Jun 06, 2025 3:41 am
Uri Blass wrote: Thu Jun 05, 2025 12:22 pm The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
In theory we could have multiple nets based on time control, with each net trained against a bot that was itself trained to simulate humans of a specified level at a specified time control, but the benefit would not justify the massive amount of both human and machine time needed. We are continuing to make progress in multiple ways. The White Queen odds was updated a few days ago to a new net that performs slightly better against bots at least, with a new net for Black likely in a week or so. The rook odds and knight odds were transferred from a 4090 based machine to a 5090 based machine which allowed for a 50% increase in max nodes with a similar average time per move; it seems to be performing even better with this change. The two knights handicap was switched to the new queen odds net, which somehow tested as more of an improvement for two knights than for queen (!), so I particularly recommend two knights odds (using "LeelaPieceOdds") for very strong players who find queen odds a bit too easy (at their preferred time control) but who hardly ever win at rook odds. Perhaps soon two knights can be a fair match in 3'2" blitz for even rather strong grandmasters. That is a sentence I would never have imagined myself ever writing a year ago! Interestingly, the low-rated GM who set off the boom in interest in LeelaQueenOdds when he lost his first five blitz (3'2") at queen odds, "Mikulas", eventually pulled ahead after hundreds of games, and made a nearly even score after that at 2'2", and is now number 9 on the Leaderboard!
The problem is that I see no way to be sure that you make progress at all time controls.

It is possible that some change that is productive at one time control is counter productive at different time control.
So far we do not know if the last knight odds bots can get 50% against GM's with playing strenght of 2550 at 90+30 time control because we have no data about games with this time control.

I know that you claim a significant improvent from the match against benjamin but when we have no data about games in this time control and even the match against benjamin is only 60+30 time control then I think that it is better to try to prove it by winning against humans with fide rating 2450-2500 before asking for a match against players above 2550 when probably 2550 will ask for significant more money than 2450-2500
lkaufman
Posts: 6227
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by lkaufman »

Father wrote: Fri Jun 06, 2025 5:29 am
lkaufman wrote: Fri Jun 06, 2025 3:41 am
Uri Blass wrote: Thu Jun 05, 2025 12:22 pm The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
In theory we could have multiple nets based on time control, with each net trained against a bot that was itself trained to simulate humans of a specified level at a specified time control, but the benefit would not justify the massive amount of both human and machine time needed. We are continuing to make progress in multiple ways. The White Queen odds was updated a few days ago to a new net that performs slightly better against bots at least, with a new net for Black likely in a week or so. The rook odds and knight odds were transferred from a 4090 based machine to a 5090 based machine which allowed for a 50% increase in max nodes with a similar average time per move; it seems to be performing even better with this change. The two knights handicap was switched to the new queen odds net, which somehow tested as more of an improvement for two knights than for queen (!), so I particularly recommend two knights odds (using "LeelaPieceOdds") for very strong players who find queen odds a bit too easy (at their preferred time control) but who hardly ever win at rook odds. Perhaps soon two knights can be a fair match in 3'2" blitz for even rather strong grandmasters. That is a sentence I would never have imagined myself ever writing a year ago! Interestingly, the low-rated GM who set off the boom in interest in LeelaQueenOdds when he lost his first five blitz (3'2") at queen odds, "Mikulas", eventually pulled ahead after hundreds of games, and made a nearly even score after that at 2'2", and is now number 9 on the Leaderboard!
Thank you, Mr. Larry Kaufman, for your teachings. I've been analyzing the new humans who are steadily rising on the LeelaQueenOdds leaderboard, and I'm placing special emphasis on one man I find "super gifted." This is GM "FeegLood." Every day, belonging to or appearing on the leaderboard of the top 100 humans becomes more difficult and meritorious. I confess that I would be very sad if my ideal horse, "Catecan," and my own self were to be forgotten. I ask God for the strength to endure on the leaderboard. I would like to be on the same level as the giants Nakamura and Magnus; then I would be ready to travel to heaven on my horse, Catecan. But truth be told, this leaderboard describes another form of greatness, a giant in itself.
GM FeegLood has been identified as Jeffery Xiong, a young American star who has been over 2700 FIDE. He has nearly a 30% score in one minute no inc bullet at queen odds, Leela white, among the best 3 or 4 records, and near-even results in blitz at rook odds with alternating colors, perhaps the best record at rook odds.
Komodo rules!
lkaufman
Posts: 6227
Joined: Sun Jan 10, 2010 6:15 am
Location: Maryland USA
Full name: Larry Kaufman

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by lkaufman »

Uri Blass wrote: Fri Jun 06, 2025 5:34 am
lkaufman wrote: Fri Jun 06, 2025 3:41 am
Uri Blass wrote: Thu Jun 05, 2025 12:22 pm The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
In theory we could have multiple nets based on time control, with each net trained against a bot that was itself trained to simulate humans of a specified level at a specified time control, but the benefit would not justify the massive amount of both human and machine time needed. We are continuing to make progress in multiple ways. The White Queen odds was updated a few days ago to a new net that performs slightly better against bots at least, with a new net for Black likely in a week or so. The rook odds and knight odds were transferred from a 4090 based machine to a 5090 based machine which allowed for a 50% increase in max nodes with a similar average time per move; it seems to be performing even better with this change. The two knights handicap was switched to the new queen odds net, which somehow tested as more of an improvement for two knights than for queen (!), so I particularly recommend two knights odds (using "LeelaPieceOdds") for very strong players who find queen odds a bit too easy (at their preferred time control) but who hardly ever win at rook odds. Perhaps soon two knights can be a fair match in 3'2" blitz for even rather strong grandmasters. That is a sentence I would never have imagined myself ever writing a year ago! Interestingly, the low-rated GM who set off the boom in interest in LeelaQueenOdds when he lost his first five blitz (3'2") at queen odds, "Mikulas", eventually pulled ahead after hundreds of games, and made a nearly even score after that at 2'2", and is now number 9 on the Leaderboard!
The problem is that I see no way to be sure that you make progress at all time controls.

It is possible that some change that is productive at one time control is counter productive at different time control.
So far we do not know if the last knight odds bots can get 50% against GM's with playing strenght of 2550 at 90+30 time control because we have no data about games with this time control.

I know that you claim a significant improvent from the match against benjamin but when we have no data about games in this time control and even the match against benjamin is only 60+30 time control then I think that it is better to try to prove it by winning against humans with fide rating 2450-2500 before asking for a match against players above 2550 when probably 2550 will ask for significant more money than 2450-2500
While time control does change typical errors, I think this is mostly in or near the bullet chess range. I don’t think there is much difference between classical games by 2600s and rapid games by 2800s, for example. I think that our next formal match will likely be rapid at rook odds with a GM. There is more interest in variety than in just repeating a previous match with slight changes. But if you know of a GM at least over 2500 who would play a classical knight odds match for small prize money, perhaps with you as intermediary between board and internet, let me know.
Komodo rules!
Kappatoo
Posts: 22
Joined: Sat Sep 23, 2017 10:33 pm

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by Kappatoo »

I don’t think there is much difference between classical games by 2600s and rapid games by 2800s, for example
Do you have data on this? I would have suspected that simple one-move blunders are much more common in the latter games.
(I'm not saying that Leela should take this into account--that might be too complex to implement to be worth it.)
Father
Posts: 1809
Joined: Sun Mar 19, 2006 4:39 am
Location: Colombia
Full name: Pablo Ignacio Restrepo

Re: Does Leela Odds bots know to play in a different way in different time controls?

Post by Father »

lkaufman wrote: Fri Jun 06, 2025 7:46 am
Father wrote: Fri Jun 06, 2025 5:29 am
lkaufman wrote: Fri Jun 06, 2025 3:41 am
Uri Blass wrote: Thu Jun 05, 2025 12:22 pm The question is if Leela Knows to play in a different way at different time controls even if the expected result is the same because of different opponents.

My guess is that if we take humans who get 50% against LeelaQueenOdds at 1+1
and humans who get 50% against LeelaQueenOdds at 10+10 then you will find that they do different types of mistakes and I guess that you are going to see more one ply blunders at 1+1 and more positinal mistakes at 10+10(did not do a research to test if I am correct).
In theory we could have multiple nets based on time control, with each net trained against a bot that was itself trained to simulate humans of a specified level at a specified time control, but the benefit would not justify the massive amount of both human and machine time needed. We are continuing to make progress in multiple ways. The White Queen odds was updated a few days ago to a new net that performs slightly better against bots at least, with a new net for Black likely in a week or so. The rook odds and knight odds were transferred from a 4090 based machine to a 5090 based machine which allowed for a 50% increase in max nodes with a similar average time per move; it seems to be performing even better with this change. The two knights handicap was switched to the new queen odds net, which somehow tested as more of an improvement for two knights than for queen (!), so I particularly recommend two knights odds (using "LeelaPieceOdds") for very strong players who find queen odds a bit too easy (at their preferred time control) but who hardly ever win at rook odds. Perhaps soon two knights can be a fair match in 3'2" blitz for even rather strong grandmasters. That is a sentence I would never have imagined myself ever writing a year ago! Interestingly, the low-rated GM who set off the boom in interest in LeelaQueenOdds when he lost his first five blitz (3'2") at queen odds, "Mikulas", eventually pulled ahead after hundreds of games, and made a nearly even score after that at 2'2", and is now number 9 on the Leaderboard!
Thank you, Mr. Larry Kaufman, for your teachings. I've been analyzing the new humans who are steadily rising on the LeelaQueenOdds leaderboard, and I'm placing special emphasis on one man I find "super gifted." This is GM "FeegLood." Every day, belonging to or appearing on the leaderboard of the top 100 humans becomes more difficult and meritorious. I confess that I would be very sad if my ideal horse, "Catecan," and my own self were to be forgotten. I ask God for the strength to endure on the leaderboard. I would like to be on the same level as the giants Nakamura and Magnus; then I would be ready to travel to heaven on my horse, Catecan. But truth be told, this leaderboard describes another form of greatness, a giant in itself.
GM FeegLood has been identified as Jeffery Xiong, a young American star who has been over 2700 FIDE. He has nearly a 30% score in one minute no inc bullet at queen odds, Leela white, among the best 3 or 4 records, and near-even results in blitz at rook odds with alternating colors, perhaps the best record at rook odds.
Good evening, Mr. Larry Kaufman.
The GM and leader of the top 100 humans facing the robot LeelaQueenOdds has crossed the 3000-rating line inherent in the leaderboard. I thought 19th place belonged to GM Maghsoodloo, Parham, currently 38th in the world. The battles to advance upwards in the leaderboard are becoming very interesting and dynamic. Today I wonder if another giant will come to challenge Arka50!! Catecan and I are resting in an old inn we found on the road. We will continue walking with a sure and steady pace like an elephant.

LeelaQueenOdds Leaderboard

Next update in: 5m 06s
# Player Rating Games Last Game TC
1 Arka50 3035 150 2025-06-06 1+0
2 SuperEasy 2940 407 2025-06-06 8+3
3 verystablegenius69 2862 384 2025-05-07 1+0
4 wateenellende 2814 154 2025-03-26 1+0
5 VladimirMarkovic 2808 1403 2025-06-06 1+0
6 InvisibleGuest2023 2795 347 2025-06-06 1+0
7 Former_Player 2778 308 2025-06-01 3+2
8 JoanFluvia 2774 631 2025-06-04 2+1
9 Mikulas 2760 596 2025-06-06 2+1
10 edgelorde 2745 1644 2025-06-06 1+1
11 Hissha 2723 135 2025-05-27 1+10
12 Catecan 2706 6502 2025-06-06 1+0
13 jflup 2705 175 2025-06-04 3+1
14 Orichess88 2694 88 2025-04-03 2+3
15 Alex_Muezersky 2673 179 2025-04-05 1+1
16 dauANHbac 2669 74 2025-04-27 1+1
17 smeets 2662 29 2025-04-02 2+1
18 Phara0h1 2662 995 2025-05-28 4+0
19 FeegLood 2626 108 2025-06-04 1+0
20 Paulewski 2574 340 2025-06-06
I am thinking chess is in a coin.Human beings for ever playing in one face.Now I am playing in the other face:"Antichess". Computers are as a fortres where owner forgot to close a little door behind. You must enter across this door.Forget the front.