CHESS BENCHMARK · LOCAL, NO THINKING

Qwen3.5 9B

Alibaba · open weights · released 2026-02-23 · route qwen3.5:9b · Model profile

Opening gap: +97 (95%: -845 to +1,039). The interval includes zero, so we can't yet say whether it performs differently across the variants.

Standard

Rating
80995%: 110 to 1,509
Record
0 won, 0 drawn, 1 lost of 1
Accuracy (open positions)
92.8
Accuracy (all moves)
92.8
Average centipawn loss
18
Blunders / mistakes / inaccuracies per 100 moves
0.0 / 0.0 / 0.0
Matched Stockfish's first choice
0%
In Stockfish's top three
100%
Legal on the first try
66.7%
Illegal / invalid
3 of 5 answers
Forfeits
1
Adjudicated games
0
API-equivalent cost per game
n/a
API-equivalent cost per move
n/a
Seconds per move
109.3
Model minutes per game
5.5
Tokens per move (in / out)
1,226 / 1,055
Reasoning tokens per move
n/a
Aborted on our side (not counted)
0

By phase

PhaseMovesAccuracyCentipawn lossBlunders /100Engine match
Opening292.8180.00%
Middlegame0n/an/an/an/a
Endgame0n/an/an/an/a

Chess960

Rating
71395%: 82 to 1,343
Record
0 won, 0 drawn, 2 lost of 2
Accuracy (open positions)
92.5
Accuracy (all moves)
92.5
Average centipawn loss
19
Blunders / mistakes / inaccuracies per 100 moves
0.0 / 0.0 / 0.0
Matched Stockfish's first choice
0%
In Stockfish's top three
0%
Legal on the first try
0.0%
Illegal / invalid
8 of 9 answers
Forfeits
2
Adjudicated games
0
API-equivalent cost per game
n/a
API-equivalent cost per move
n/a
Seconds per move
281.9
Model minutes per game
7.0
Tokens per move (in / out)
2,133 / 2,512
Reasoning tokens per move
n/a
Aborted on our side (not counted)
0

By phase

PhaseMovesAccuracyCentipawn lossBlunders /100Engine match
Opening192.5190.00%
Middlegame0n/an/an/an/a
Endgame0n/an/an/an/a

Games