Models

Model set

Every model with an official Overall rank is shown by default. Partial results keep their rank; the icon explains what evidence is incomplete.

Reasoning view

Rank by

Reasoning view applies to Scores, Reasoning variants, Head-to-head, Per game heatmap, Economics (when available), and Cycle detector.

Compare two entries

Select two models to inspect their evidence together.

Rankings

Selected models, ordered by the Rank by choice above

Conservative score is the public leaderboard score after public-score adjustments; the other Rank by choices show diagnostic views when the selected models have that data.

Scores

Overall min-max and mean; small dots are each game (averaged across reasoning variants where scores exist)

Economics

Coming soon

Generation cost versus score will appear when the economics segment is available.

Overall

Reasoning variants

Per game

Strength heatmap (per-game scores averaged across reasoning variants)

Head-to-head

Combined view across tested provider settings.

Off-diagonal cells use win rate for the row model against the column opponent (W-L-D in the tooltip). Use Reasoning view to show one reasoning variant.

More charts

Less-used views open on demand, so the page stays short.

Per-game comparison Top selected models within each game

Each panel ranks selected models within one game for the active reasoning view.

Cycle detector Non-transitive matchups between selected models

What it finds: Leaderboard averages can hide matchup loops. The detector searches the currently selected models and reasoning view for cases where A outperforms B, B outperforms C, and C outperforms A.

How to read it: In A → B → C → A, each arrow points from the model with the higher pair score to the model it outscored. Hover over a result for each edge’s win-loss-draw record and pair score.

Each arrow requires at least four matches and a pair score above 50%, with a draw counting as half a win. Results are ordered by the cycle’s weakest edge: weakest edge +x pp is the smallest of the three margins above 50%, while n is the total match count across all three edges. The detector shows up to ten cycles. These are descriptive matchup patterns, not tests of statistical significance. “No cycles” means none met these rules in the current selection and reasoning view; it does not prove that the matchups are transitive.

Output tokens and API latency How much each model wrote, and how fast

These views will appear when the economics segment is available.

Static rankings — current page release

This table remains available without chart scripts. It describes this page's release; restoring a requested comparison and its filters requires JavaScript. Open benchmark rankings for the detailed tables.

Current comparison tables

Overall leaderboard scores

All officially ranked Overall models from the same data as the interactive charts. Partial rows keep their official rank and show that their evidence is incomplete.

Model	Evidence status	Rank	Mean	Min	Max
GPT-6 Astra	Complete evidence	1	79.4	38.6	100.0
Claude Opus 5.5	Complete evidence	2	77.6	29.1	100.0
Claude Fable 5.1	Complete evidence	3	76.1	36.4	99.6
GPT-6.1 Sol	Complete evidence	4	73.6	30.8	100.0
Claude Opus 5	Complete evidence	5	59.1	14.8	92.9
GPT-6 Sol	Complete evidence	6	58.2	11.4	88.6
Grok 4.7	Complete evidence	7	57.1	12.3	89.4
GLM-5.3	Coverage incomplete: the official generation cohort is incomplete	8	53.6	10.0	91.7
GPT-5.6 Sol	Complete evidence	9	51.0	11.3	80.2
Claude Fable 5	Complete evidence	10	49.8	6.8	81.5
Claude Sonnet 5.5	Complete evidence	11	49.4	7.4	100.0
MiMo-V2.6-Pro	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing; the official generation cohort is incomplete	12	47.6	7.5	77.3
Grok 4.6	Complete evidence	13	43.8	9.7	74.2
GPT-5.4	Complete evidence	14	43.4	2.2	100.0
Gemini 3.8 Flash	Complete evidence	15	43.0	5.6	79.2
Muse Spark 1.3	Coverage incomplete: the official generation cohort is incomplete	16	42.6	10.9	79.4
GLM-5.3 Flash	Complete evidence	17	41.4	10.1	74.8
Grok 4.5	Complete evidence	18	41.0	0.0	73.0
GLM-5.3 FlashX	Complete evidence	19	40.8	7.2	74.5
Claude Opus 4.8	Complete evidence	20	40.6	6.3	78.6
Hy4 Preview	Complete evidence	21	40.3	3.5	74.9
Qwen3.8 Max 0902	Coverage incomplete: the official generation cohort is incomplete	22	39.9	4.9	81.0
Qwen3.8 2.4t A95B	Complete evidence	23	38.5	6.6	75.5
Ox Alpha	Complete evidence	24	38.5	9.3	65.9
MiMo-V2.6-Flash	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	25	38.2	8.5	72.3
GPT-4.1 Nano	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing; the official generation cohort is incomplete	26	38.1	38.1	38.1
Gemini 3.6 Flash	Complete evidence	27	37.8	4.3	77.5
DeepSeek V4.1 Flash	Complete evidence	28	37.5	7.6	77.4
GPT-5.5	Complete evidence	29	37.3	5.1	77.7
Ember-1	Complete evidence	30	36.7	8.6	66.4
Gemini 3.5 Flash	Coverage incomplete: fixed opponent-and-seat coverage is incomplete	31	36.6	5.8	78.7
Kimi K3	Coverage incomplete: the official generation cohort is incomplete	32	35.3	7.6	78.1
Muse Spark 1.2	Coverage incomplete: the official generation cohort is incomplete	33	35.2	7.0	68.0
Qwen3.8 Max	Complete evidence	34	34.9	7.2	81.8
Claude Opus 4.5	Complete evidence	35	34.4	9.0	57.0
Claude Sonnet 5	Complete evidence	36	33.1	8.3	58.5
Qwen3.8 Flash	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	37	33.0	8.7	73.7
GPT-5.6 Terra	Complete evidence	38	32.8	10.2	73.6
DeepSeek V4 Pro	Complete evidence	39	32.7	6.0	72.9
Gemini 3.1 Pro Preview	Complete evidence	40	32.7	8.5	74.1
DeepSeek V4 Pro 0813	Complete evidence	41	32.4	1.3	68.5
GPT-6 Luna	Coverage incomplete: the official generation cohort is incomplete	42	32.1	9.7	79.9
Qwen3.8 27B	Coverage incomplete: generation quality is below the publication threshold; fixed opponent-and-seat coverage is incomplete; the official generation cohort is incomplete	43	30.8	14.6	66.2
GPT-5	Coverage incomplete: one or more documented settings are untested or ineligible; fixed opponent-and-seat coverage is incomplete	44	30.3	0.0	68.1
Qwen3.7 Max	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	45	30.2	9.7	65.7
GLM-5.2	Complete evidence	46	30.1	7.3	73.3
MiMo-V2.5-Pro	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	47	29.8	5.9	65.4
o3	Coverage incomplete: generation quality is below the publication threshold	48	29.8	9.5	52.1
Kimi K2.7 Code	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	49	28.4	8.4	47.2
GPT-5.6 Luna	Complete evidence	50	28.4	0.2	75.8
GPT-4.1	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	51	27.7	6.8	50.1
Space Bunny Alpha	Complete evidence	52	26.9	8.1	64.4
Seed 2.0 Code	Complete evidence	53	26.7	5.8	68.2
GPT-5.4 Mini	Coverage incomplete: one or more bracket scores are missing; one or more documented settings are untested or ineligible; generation-quality evidence is missing	54	26.4	7.9	54.1
GPT-5.4 Nano	Coverage incomplete: one or more bracket scores are missing; one or more documented settings are untested or ineligible; generation-quality evidence is missing	55	25.8	6.6	56.7
DeepSeek V4 Flash	Complete evidence	56	25.3	6.9	75.4
Gemini 3.5 Flash Lite	Complete evidence	57	24.8	6.9	55.5
Qwen3.7 Plus	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	58	24.6	6.1	67.6
Laguna S 2.1	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	59	24.2	6.2	45.1
Hy3 Preview	Coverage incomplete: one or more bracket scores are missing; one or more documented settings are untested or ineligible; generation-quality evidence is missing	60	24.2	2.7	60.8
Grok Build 0.1	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	61	24.1	9.9	46.9
Step 3.7 Flash	Coverage incomplete: generation quality is below the publication threshold	62	23.5	9.7	46.6
Gemma 4 31B	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing; the official generation cohort is incomplete	63	23.1	12.2	45.2
GPT-OSS 120B	Complete evidence	64	22.3	8.1	52.1
LongCat 2.0	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	65	22.3	6.2	42.6
Muse Glimmer 30B	Complete evidence	66	22.1	8.2	43.9
Nemotron 3 Ultra 550B A55B	Coverage incomplete: one or more documented settings are untested or ineligible	67	21.9	4.4	56.0
Inkling Small	Complete evidence	68	21.7	5.4	52.6
Minimax M3	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	69	21.3	0.0	45.3
Inkling	Coverage incomplete: generation quality is below the publication threshold	70	20.8	8.1	51.2
MiMo-V2.5	Coverage incomplete: three distinct provider settings are not documented; one or more documented settings are untested or ineligible	71	20.4	6.6	48.8
North Mini Code	Coverage incomplete: one or more bracket scores are missing; the provider capability record is unavailable; generation-quality evidence is missing; generation quality is below the publication threshold; fixed opponent-and-seat coverage is incomplete	72	17.8	4.9	44.0
Mercury 2.5 Preview	Complete evidence	73	17.4	8.1	38.8
Mistral Medium 3.5	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing	74	17.3	0.0	40.2
Nex N2 Pro	Coverage incomplete: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing; generation quality is below the publication threshold	75	16.9	0.0	56.4
Gemini 3.1 Flash Lite	Coverage incomplete: one or more documented settings are untested or ineligible	76	16.8	0.0	35.7
Ling 3.0 Flash	Coverage incomplete: one or more bracket scores are missing; the provider capability record is unavailable; generation-quality evidence is missing; generation quality is below the publication threshold; fixed opponent-and-seat coverage is incomplete	77	14.7	1.8	36.9
Granite 4.2 8B	Coverage incomplete: one or more bracket scores are missing; one or more documented settings are untested or ineligible; generation-quality evidence is missing; generation quality is below the publication threshold	78	8.1	5.1	15.7

Charts

Chart controls