🤖 AI benchmark: hit-rate of 7 models

Prematch Na żywo (in-play)

Seven external AI models (Hermes contour) independently analyze the same mecze — predicting the wynik (1X2), total (Powyżej/Poniżej), both drużyny to wynik (BTTS) and the exact wynik. Here we honestly compare their typy against the real result after the final whistle and combine everything into a single accuracy rating. An informational and analytical snapshot, not betting advice.

⚠️ Data is still accumulating — counting starts from 09.07.2026, so all models are compared on the same events (early test typy are excluded). The sample is still small and not representative. Right now the snapshot holds 2683 mecz(es), 9049 settled AI typy (Piłka nożna, Tenis, Koszykówka, Siatkówka, Hokej, Tenis stołowy). The figures below are N, not «a percentage you can trust»: the more mecze are played out, the more reliable the snapshot becomes. We show it transparently from day one, not only once the sample becomes «convenient».

Wszystkie sporty Piłka nożna Tenis Koszykówka Siatkówka Hokej Tenis stołowy

Leaderboard

Model	N (settled)	1X2	Podwójna szansa (1X)	Exact wynik	Composite accuracy
ChatGPT	263	65.0%(171/263)	74.6%(47/63)	17.2%(35/203)	44.2%(206/466)
Claude	2170	61.4%(1332/2170)	77.6%(612/789)	21.9%(414/1891)	43.0%(1746/4061)
Kimi	1109	60.8%(674/1109)	73.8%(253/343)	20.3%(196/964)	42.0%(870/2073)
Google AI	2810	60.3%(1694/2808)	74.7%(735/984)	20.0%(554/2768)	40.3%(2248/5576)
GLM 5.2	1285	59.5%(765/1285)	74.4%(302/406)	19.5%(246/1261)	39.7%(1011/2546)
DeepSeek	175	60.6%(106/175)	56.1%(23/41)	18.5%(32/173)	39.7%(138/348)
Qwen	1237	59.3%(733/1237)	75.0%(267/356)	16.8%(196/1166)	38.7%(929/2403)

grey — sample <5, not representative; «—» — the model has not made a settled typ yet.

Podwójna szansa (1X) — the same typ counts as a wygrana if the chosen side won or the mecz drew. Of the 1X2 losses in football/hockey: 680 draws, 743 underdog (total settled 1X2 typy in these sports: 2982, double chance combined 75.1% (2239/2982)). The models almost always take the favorite and don't bet on a remis — double chance shows how many bets are eaten specifically by draws.

Composite accuracy — the share of correct typy w all pokazano rynki together: (sum of correct typy) ÷ (sum of all settled typy) w the rynki 1X2 + Exact wynik. Each rynek-typ weighs equally. This is hit-rate, not profitability — for money/ROI by model see /ai-agent. «Exact wynik» — the full final wynik was guessed correctly (H and A matched); for tennis, sets are compared; typy with no recognized wynik do not count toward the denominator. Podwójna szansa (1X): a typ counts as a wygrana if the chosen side won OR it was a remis — it accounts for frequent draws that «eat» bets on the favorite. This metric is informational and is not included in composite accuracy. On «Wszystkie sporty» we don't show total and BTTS — they're tied to the sport and make no sense in a mixed pool.

Composite model rating · all rynki

Bar height = the model's composite accuracy w all applicable rynki on the current sample. Sorted from best to worst.

44.2% (206/466)

GPT 5.5

43.0% (1746/4061)

Opus 4.8

42.0% (870/2073)

Kimi 2.6

40.3% (2248/5576)

Gemini 3.5 Flash

39.7% (1011/2546)

GLM 5.2

39.7% (138/348)

DeepSeek V4 Pro

38.7% (929/2403)

Qwen 3.7 Plus

Bars are AI models by version; grey/dimmed — sample <5, not representative. The snapshot is informational, not betting advice.

Accuracy by rynek

Where each model is strong: one mini-bar per applicable rynek, with the percentage and (hits/sample).

ChatGPT Composite 44.2%

1X2

65.0% (171/263)

Exact wynik

17.2% (35/203)

Claude Composite 43.0%

1X2

61.4% (1332/2170)

Exact wynik

21.9% (414/1891)

Kimi Composite 42.0%

1X2

60.8% (674/1109)

Exact wynik

20.3% (196/964)

Google AI Composite 40.3%

1X2

60.3% (1694/2808)

Exact wynik

20.0% (554/2768)

GLM 5.2 Composite 39.7%

1X2

59.5% (765/1285)

Exact wynik

19.5% (246/1261)

DeepSeek Composite 39.7%

1X2

60.6% (106/175)

Exact wynik

18.5% (32/173)

Qwen Composite 38.7%

1X2

59.3% (733/1237)

Exact wynik

16.8% (196/1166)

The model's favorite by 1X2 = the max of P1/X/P2 in its probabilities; for sports without a remis (tennis, volleyball, etc.) the «X» option doesn't participate. grey — sample <5, not representative. Not betting advice.