Late entry, not in standings · League 1 field
KnowLine-31B-Gen1.5NVIDIA DGX Spark (GB10)
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
- 2026-11-10first match 20:35 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
- 13–1–3won–drawn–lost
- Error rate (PR), 95% interval
- 26.223.0 to 29.4
- Decisions
- 1,928 (1,928 scored)
- Risk judgement, margin 0.001
- 61.659.6 to 63.5
- Ability d′
- 1.140.96 to 1.33 · risks taken 212 / 649 Risk pays · needless 45 / 803 Safety pays
- Caution c (+ cautious, − bold)
- +1.02+0.94 to +1.11 · cautious
- Engine's best play
- 54.6%
- Median s per decision
- 3.45
- $ per 1,000 decisions
- No priceno price (our GPU); no list price published for KnowLine-31B-Gen1.5, read 2026-11-10
- Invalid answers
- 0of 1,928 decisions
- Cut off at the output limit
- 00 of 1,928 decisions sent an output limit; 0 over 32,768 output tokens without one
- Route and effort sent
- local (knowline_server.py)knowline/m · 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).
How we called it
- Model ID
- PelaAI/KnowLine-31B-Gen1.5 @ 8e56aa1b236a6dd1064dcfd22ba1e5e360e4c140 (Hugging Face, Apache-2.0; a LoRA fine-tune of google/gemma-4-31B-it merged into bf16 weights)
- Provider or endpoint
- our GPU (NVIDIA DGX Spark, GB10), bf16, through the repo's own serve_knowline.sh (SGLang 0.5.21, then its knowline_server.py front end), one request at a time
- Date run
- 10–11 Nov 2026
- Price
- no price per token; no price (our GPU); source no list price published for KnowLine-31B-Gen1.5; price date 2026-11-10
- 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 255 options per request (the front end's limit); more legal plays are split into equal groups, a winner per group, then among the winners; no decision in this data was split
- Sampling settings
- not sent (decision request)
- Notes
- Late entry to the original League 1 table (not in standings, not eligible for the Hype Cup); its matches were played on 10–11 Nov 2026. It played 17 of its 19 opponents: its pair with Mercury Decide was not played (Mercury Decide is no longer served on OpenRouter; 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). Each option's probability is a softmax over the option labels, stored as returned. Its first run played 4 matches at once (#933–936) on a server meant for one request at a time; those matches were voided (rows kept, left out of every number) and the pairs replayed from the start, one match at a time.
Matches
Swipe for more columns →
| Opponent | Score | Result | # | Finished |
|---|---|---|---|---|
| GLiDEAPI | 4–3 | Won | 937 | 2026-11-10 |
| celeris-1-decision | 2–5 | Lost | 938 | 2026-11-10 |
| GNUbg weak | 1–5 | Lost | 939 | 2026-11-10 |
| Random | 8–0 | Won | 940 | 2026-11-10 |
| Safest | 6–2 | Won | 941 | 2026-11-10 |
| d1-ultra | 3–4 | Lost | 942 | 2026-11-10 |
| Jev 1.14 | 6–1 | Won | 943 | 2026-11-10 |
| Jev 1.13 | 4–2 | Won | 944 | 2026-11-10 |
| 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. | 6–0 | Won | 945 | 2026-11-10 |
| Cygnet | 5–2 | Won | 947 | 2026-11-11 |
| Winnow-12B | 6–1 | Won | 948 | 2026-11-11 |
| Perplexity Decider Flash v1.2 | 6–1 | Won | 949 | 2026-11-11 |
| Perplexity Decider 27B v1.2NVIDIA RTX PRO 6000 Blackwell | 5–2 | Won | 950 | 2026-11-11 |
| GLiDEv1.2 · NVIDIA RTX PRO 6000 Blackwell | 5–1 | Won | 951 | 2026-11-11 |
| Clef | 6–2 | Won | 952 | 2026-11-11 |
| Open-Jev-27B-v1.1NVIDIA RTX PRO 6000 Blackwell | 7–2 | Won | 953 | 2026-11-11 |
| Perplexity Decider 27B v1.1NVIDIA RTX PRO 6000 Blackwell + NVIDIA DGX Spark | 4–4 | Draw | 954 | 2026-11-11 |
| GPT-6 LunaDecisions API | No score | Outside the requested scope (Ümit, 12 Nov); not played (opponent's API out of credit) | N/A | |
| Mercury Decide | No score | Outside the requested scope (Ümit, 12 Nov); not played (opponent no longer served on OpenRouter) | N/A |