Classifica

Classifica: Matematica

Dalla matematica olimpica alla frontiera: AIME, MATH, FrontierMath.

Generale700 Coding399 Matematica53 Reasoning413 Agentico55 Conoscenza412 Multimodale382 Agenti di codice55 Reasoning scientifico82 Multilingue12 Scrittura61 Chat e assistenti61 Ricerca85 Chiamata di strumenti32 Contesto lungo82 Velocità402 Valore401 Open weights170

53 modelli

#	Modello	Podium Score	AIME 2025	FrontierMath	MATH	Velocità
01	Claude Mythos Preview Anthropic · reasoning · preview	100.0	—	—	—	80 t/s
02	MiMo V2.5 Pro Xiaomi · Aperto	100.0	—	—	86.2%	65 t/s
03	GPT-6 Astra OpenAI · reasoning	97.6	—	97.6%	—	82 t/s
4	Grok 4 Heavy xAI · preview	84.6	100%	—	—	50 t/s
5	Qwen3.7 Max Alibaba	81.6	—	—	—	203 t/s
6	Claude Opus 4.6 Anthropic	79.6	99.8%	—	—	50 t/s
7	Gemini 3.1 Pro Google	76.0	—	—	—	136 t/s
8	Gemini 3 Pro Google	72.9	100%	—	—	120 t/s
9	Kimi K3 Moonshot AI · Aperto · reasoning	70.9	—	—	—	37 t/s
10	GPT-5.6 Sol OpenAI	70.7	—	89%	—	72 t/s
11	GLM-5.2 Z.ai · Aperto	69.8	—	—	—	164 t/s
12	Gemini 3 Flash Google	69.6	99.7%	—	—	120 t/s
13	Qwen3.8 Flash Next Alibaba · Aperto	68.0	—	—	—	168 t/s
14	DeepSeek V4.1 Flash DeepSeek · Aperto · reasoning	68.0	—	—	—	214 t/s
15	DeepSeek V4 Pro DeepSeek · Aperto	66.5	—	—	—	66 t/s
16	GLM-5.3 Flash Z.ai · Aperto	64.0	—	—	—	182 t/s
17	Qwen3.7 Plus Alibaba	62.0	—	—	—	54 t/s
18	DeepSeek V4 Flash DeepSeek · Aperto	62.0	—	—	—	113 t/s
19	Muse Spark Meta	59.8	—	—	—	80 t/s
20	Qwen3.6 Plus Alibaba	59.2	—	—	—	55 t/s
21	Claude Opus 4.8 Anthropic	58.7	—	—	—	50 t/s
22	GPT-5.6 Terra OpenAI	57.3	—	84.9%	—	127 t/s
23	Claude Opus 4.7 Anthropic	57.0	—	—	—	80 t/s
24	GPT-5.2 OpenAI	53.0	100%	40.3%	—	90 t/s
25	Seed 2.0 Pro ByteDance	47.2	98.3%	—	—	40 t/s
26	Kimi K2.6 Moonshot AI · Aperto	46.9	—	—	—	40 t/s
27	Qwen3.5 397B-A17B Alibaba · Aperto	45.3	—	—	—	71 t/s
28	Agnes 3.0 Flash Agnes · reasoning	42.0	—	—	—	235 t/s
29	GPT-5.6 Luna OpenAI	40.3	—	78.6%	—	175 t/s
30	DeepSeek V4 Flash Vision DeepSeek	39.0	—	—	—	120 t/s
31	G9v3-39A5B AI9Stars · Aperto	39.0	—	—	—	60 t/s
32	G9v3-3B AI9Stars · Aperto	39.0	—	—	—	60 t/s
33	GLM-5.3 Zhipu AI	39.0	—	—	—	90 t/s
34	Grok 4.6 (low) xAI	39.0	—	—	—	59 t/s
35	Grok 4.6 (medium) xAI	39.0	—	—	—	59 t/s
36	Grok 4.6 (xhigh) xAI	39.0	—	—	—	62 t/s
37	HyperNova 60B Multiverse · Aperto	39.0	—	—	—	367 t/s
38	LFM2.5-2.6B Liquid AI · Aperto	39.0	—	—	—	60 t/s
39	LFM2.5-8B-A1B Liquid AI · Aperto	39.0	—	—	—	338 t/s
40	LongCat 2.0 LongCat · Aperto	39.0	—	—	—	42 t/s
41	Magistral Small 1.2 Mistral · Aperto	39.0	—	—	—	60 t/s
42	Qwen3.8 27B (xhigh) Alibaba · Aperto	39.0	—	—	—	53 t/s
43	Qwen3.8 27B (low) Alibaba · Aperto	39.0	—	—	—	64 t/s
44	Qwen3.8 27B (medium) Alibaba · Aperto	39.0	—	—	—	62 t/s
45	Qwen3.8 27B (Non-reasoning) Alibaba · Aperto · reasoning	39.0	—	—	—	62 t/s
46	Hunyuan Hy3 Tencent	35.8	—	—	—	69 t/s
47	GPT-5.4 OpenAI	32.1	—	47.6%	—	60 t/s
48	Claude Sonnet 4.6 Anthropic	30.7	—	—	—	50 t/s
49	GPT-5.5 OpenAI	25.7	—	35.4%	—	60 t/s
50	GPT-5.1 OpenAI	13.0	94%	—	—	60 t/s
51	GPT-5 OpenAI	10.9	94.6%	—	84.7%	60 t/s
52	Qwen3.8 Max Alibaba	5.0	—	—	—	55 t/s
53	DeepSeek V3.2 DeepSeek · Aperto	0.0	93.1%	—	—	60 t/s

Il Podium Score è una combinazione ponderata di cinque classifiche. Consulta la Metodologia per i dettagli.