Late entry, not in standings · League 1 field

GLiClass v4 Edge

Published

Results f24d1f41, updated 16 November 2026, 21:24 UTC; scored by GNU Backgammon 1.08, 2-ply.

Late entry, not in standings

Late entry
Late entry, not in standings: its games count in no standings table, no league table, no Hype Cup input and no other setup's numbers
Entry date
2026-11-12first match 08:43 UTC
League field played
League 1 (original table)19 setups
Matches played
17 / 19GPT-6 LunaDecisions API: outside the requested scope (Ümit, 12 Nov); not played (opponent's API out of credit)Mercury Decide: outside the requested scope (Ümit, 12 Nov); not played (opponent no longer served on OpenRouter)
Record
0–0–17won–drawn–lost
Error rate (PR), 95% interval
133.5118.5 to 148.8
Decisions
1,405 (1,405 scored)
Risk judgement, margin 0.001
21.518.6 to 24.6
Ability d′
0.340.13 to 0.57 · risks taken 461 / 508 Risk pays · needless 400 / 478 Safety pays
Caution c (+ cautious, − bold)
−1.15−1.29 to −1.02 · bold
Engine's best play
15.7%
Median s per decision
1.33
$ per 1,000 decisions
0.174estimate · DGX Spark
Invalid answers
0of 1,405 decisions
Cut off at the output limit
00 of 1,405 decisions sent an output limit; 0 over 32,768 output tokens without one
Route and effort sent
local (gliclass_server.py)gliclass · effort not in the requests

A late entry joins an original league table after it was played and plays every setup of that table once, in matches of its own. Its games count in no standings table, no league table, no Hype Cup input and no other setup's numbers. The numbers here come from its own late-entry games only, computed as for a setup in a league.

Error rate and risk judgement: 95% intervals, whole games resampled. Ability d′ and caution c at the 0.001 margin: rates corrected by adding 0.5 to the counts and 1 to the totals; 95% intervals, blocks of two seat-swapped games resampled. Cut off: answers that stopped at an output limit the request sent (finish reason length, stop reason max_tokens, finishReason MAX_TOKENS, incomplete max_output_tokens), in the stored response or error. Effort: the reasoning effort the requests sent. Engine's best play: the share of its scored decisions that chose the engine's best play (equity loss under the best-play threshold). Time: median request seconds per decision (waits between retries not included). Cost per 1,000 decisions: the recorded tokens at the price shown, as labelled there (list or billed price; for a setup on our GPUs the hosted model's list price, not billed).

Late entries on the Leagues page · JSON

How we called it

Model ID
knowledgator/gliclass-v4-edge @ 885af1af8a07edf4ed34df83ca0d995ed8201283 (Hugging Face, Apache-2.0)
Provider or endpoint
our GPU (NVIDIA DGX Spark, GB10, spark2), BF16, through turkce's gliclass_server.py with the card's own System One mapping (gliclass 0.1.22, one pass per choice question), one request at a time
Date run
12 and 14 Nov 2026
Price
no price per token; estimate: GPU time, at most; source content/own_cost.json (wall clock on our machines × hourly rate); price date not recorded
Model type
decision model
Weights
open
Runs on
our GPUs
Reasoning setting
none (decision model)
Confidence source
decision-model probability
Rules text
long
Option groups
up to 218 options per request (its 16,384-token input); more legal plays are split into equal groups, a winner per group, then among the winners; in this data 19 decisions were split into groups of 113, 120, 124, 127, 129, 132, 135, 139, 159, 160, 165, 178, 198, 218
Sampling settings
not sent (decision request)
Notes
Late entry to the original League 1 table (not in standings, not eligible for the Hype Cup), late-entry league #36. knowledgator/gliclass-v4-edge @ 885af1a on our DGX Spark (spark2, the whole GB10, handed over by logic one server at a time), BF16, through turkce's gliclass_server.py with the card's own System One mapping (gliclass 0.1.22, one pass per choice question), one request at a time; never on a hosted API. First stage 12 Nov 2026: its 9 hosted and program opponents (#1141–#1169, runner e4ce0e9); local stage 14 Nov 2026: the 8 setups we run ourselves, one at a time on an RTX PRO 6000 Blackwell of ours (#1363–#1370, runner 2a22d1f). It played 17 of its 19 opponents: its pair with Mercury Decide was not played (Mercury Decide is no longer served on OpenRouter; match #1170 aborted at its warm-up; outside the requested scope, Ümit, 12 Nov) and its pair with GPT-6 Luna was not played (GPT-6 Luna's API was out of credit; outside the requested scope, Ümit, 12 Nov). Options, instructions and state share one 16,384-token input (refused with 422 above it, never cut); League 1's worst case, the longest state plus 218 options at the longest label, is 16,367 tokens, so play lists over 218 options are split into equal groups of at most 218 and a final among the group winners (the league's split rule, as Winnow and Clef; 24 of League 1's 36,995 decisions have more than 218 options). Nothing shortened or refused. Its first stage was interrupted once (12 Nov 08:59 UTC, the server failed 3 health checks in a row) and resumed at 10:15 UTC. Its cost is an estimate from its time on the DGX Spark (GPU time, at most).

Matches

Swipe for more columns →

GLiClass v4 Edge against each opponent, its score first: 0 won, 0 drawn, 17 lost; 2 outside the requested scope (Ümit, 12 Nov), not played, not a loss; none counts in a standings table
OpponentScoreResult#Finished
GLiDEAPI0–8Lost11412026-11-12
celeris-1-decision1–8Lost11422026-11-12
GNUbg weak0–5Lost11432026-11-12
Random1–5Lost11442026-11-12
Safest0–7Lost11652026-11-12
d1-ultra0–7Lost11662026-11-12
Jev 1.140–6Lost11672026-11-12
Jev 1.130–6Lost11682026-11-12
SubQ Snap PreviewThe opponent was requested as SubQ Snap Preview; under that name the API returned answers identical to SubQ Snap 1.1 on 1,000 paired positions, so the opponent's model identity is not verified.0–7Lost11692026-11-12
Cygnet0–5Lost13632026-11-14
Winnow-12B1–6Lost13642026-11-14
Perplexity Decider Flash v1.20–6Lost13652026-11-14
Perplexity Decider 27B v1.2NVIDIA RTX PRO 6000 Blackwell1–5Lost13662026-11-14
GLiDEv1.2 · NVIDIA RTX PRO 6000 Blackwell0–7Lost13672026-11-14
Clef0–8Lost13682026-11-14
Open-Jev-27B-v1.1NVIDIA RTX PRO 6000 Blackwell0–7Lost13692026-11-14
Perplexity Decider 27B v1.1NVIDIA RTX PRO 6000 Blackwell + NVIDIA DGX Spark0–5Lost13702026-11-14
GPT-6 LunaDecisions APINo scoreOutside the requested scope (Ümit, 12 Nov); not played (opponent's API out of credit)N/A
Mercury DecideNo scoreOutside the requested scope (Ümit, 12 Nov); not played (opponent no longer served on OpenRouter)N/A