Classifica

Classifica: Open weights

Modelli open weights per Podium Score.

Generale767 Coding469 Matematica90 Reasoning483 Agentico89 Conoscenza482 Multimodale452 Agenti di codice68 Reasoning scientifico96 Multilingue12 Scrittura95 Chat e assistenti95 Ricerca99 Chiamata di strumenti45 Contesto lungo116 Velocità472 Valore471 Open weights183

183 modelli

#	Modello	Podium Score	Arena Elo	Intelligenza	Programmazione	Velocità	Prezzo (uscita)
01	Kimi K3.1 Pro Moonshot AIApertoreasoning	94.2	1490	68.0	72.0	45 t/s	$20/M
02	Kimi K3.1 Highspeed Moonshot AIApertoreasoning	91.5	1455	60.0	60.0	130 t/s	$8/M
03	FLUX.2 Dev Black Forest LabsAperto	90.4	1440	90.0	0.0	28 t/s	$0.03/M
4	Qwen Audio 3.1 TTS+ AlibabaAperto	89.4	1430	89.0	0.0	140 t/s	$0.02/M
5	GLM-4.5 Air Zhipu AIApertoreasoning	88.9	1420	58.0	55.0	96 t/s	$1.1/M
6	MiMo v2.6 Flash XiaomiAperto	88.4	1425	88.0	86.0	118 t/s	$0.3/M
7	FLUX.2 Klein 9B Black Forest LabsAperto	88.2	1405	88.0	0.0	55 t/s	$0.02/M
8	NVIDIA Nemotron 3.5 Super-VL NVIDIAAperto	87.8	1418	88.0	83.0	110 t/s	$1.2/M
9	Wan 2.2 A14B AlibabaAperto	87.4	1415	87.0	0.0	15 t/s	$0.06/M
10	FLUX.2 Klein 4B Black Forest LabsAperto	86.5	1380	87.0	0.0	90 t/s	$0.01/M
11	Kimi K3 Moonshot AIApertoreasoning	83.3	1485	57.0	61.5	37 t/s	$15/M
12	Qwen3.8 Flash Next AlibabaAperto	66.8	1460	61.5	60.5	168 t/s	$0.4/M
13	G9v3-39A5B AI9StarsAperto	65.0	—	30.0	—	60 t/s	$2/M
14	G9v3-3B AI9StarsAperto	65.0	—	30.0	—	60 t/s	$2/M
15	HyperNova 60B MultiverseAperto	65.0	—	30.0	—	367 t/s	$0.14/M
16	LFM2.5-2.6B Liquid AIAperto	65.0	—	30.0	—	60 t/s	$2/M
17	LFM2.5-8B-A1B Liquid AIAperto	65.0	—	30.0	—	338 t/s	$2/M
18	LFM2.5-VL-1.6B Liquid AIAperto	65.0	—	30.0	—	395 t/s	$2/M
19	LongCat 2.0 LongCatAperto	65.0	—	30.0	—	42 t/s	$2.95/M
20	Magistral Small 1.2 MistralAperto	65.0	—	30.0	—	60 t/s	$1.5/M
21	Qwen3.8 27B (xhigh) AlibabaAperto	65.0	—	30.0	—	53 t/s	$3/M
22	Qwen3.8 27B (low) AlibabaAperto	65.0	—	30.0	—	64 t/s	$3/M
23	Qwen3.8 27B (medium) AlibabaAperto	65.0	—	30.0	—	62 t/s	$3/M
24	Qwen3.8 27B (Non-reasoning) AlibabaApertoreasoning	65.0	—	30.0	—	62 t/s	$3/M
25	GLM-5.3 Flash Z.aiAperto	64.2	1445	58.0	58.4	182 t/s	$0.3/M
26	GLM-5.2 Z.aiAperto	59.1	1469	51.0	41.9	164 t/s	$4.29/M
27	Granite 4.2 30B Instruct IBMAperto	58.6	1380	51.0	56.0	78 t/s	$1.2/M
28	Qwen3 8 2 4t A95b AlibabaAperto	57.7	—	—	40.0	85 t/s	$2/M
29	DeepSeek V4.1 Flash DeepSeekApertoreasoning	54.5	1450	40.0	52.0	214 t/s	$1.2/M
30	Glm 5 Z.aiAperto	52.6	—	—	40.0	60 t/s	$3/M
31	MiMo V2.6 Pro XiaomiAperto	52.5	—	—	45.0	72 t/s	$0.95/M
32	Sarvam 105B Sarvam AIAperto	50.0	—	—	—	60 t/s	$3/M
33	Sarvam 30B Sarvam AIAperto	50.0	—	—	—	60 t/s	$3/M
34	DeepSeek V4.1 Flash (Non-reasoning) DeepSeekAperto	50.0	—	—	45.0	240 t/s	$1.2/M
35	Granite 4.2 8B Instruct IBMAperto	49.2	1315	43.0	44.0	115 t/s	$0.45/M
36	Kimi K2.6 Moonshot AIAperto	49.0	1461	—	36.4	40 t/s	$4/M
37	Sarvam M Reasoning Sarvam AIApertoreasoning	49.0	—	—	—	40 t/s	$3/M
38	MiMo V2.5 Pro XiaomiAperto	48.0	1466	42.0	42.6	65 t/s	$0.87/M
39	DeepSeek V4 Pro DeepSeekAperto	46.0	1458	44.0	30.7	66 t/s	$0.87/M
40	GLM-5.1 Z.aiAperto	46.0	—	—	57.0	74 t/s	$4.4/M
41	DeepSeek V4 Flash 0420 (Non-reasoning) DeepSeekAperto	46.0	—	—	38.0	210 t/s	$1.1/M
42	Deepseek V4 Pro 0424 DeepSeekAperto	45.3	—	—	40.0	60 t/s	$1.1/M
43	Deepseek V4 Pro 0424 High DeepSeekAperto	43.7	—	—	40.0	60 t/s	$1.1/M
44	Deepseek V4 Flash 0420 DeepSeekAperto	42.1	—	—	25.0	60 t/s	$1.1/M
45	DeepSeek V4 Flash DeepSeekAperto	42.0	1438	50.0	38.0	113 t/s	$0.28/M
46	Granite 4.2 3B Instruct IBMAperto	42.0	1260	36.0	35.0	148 t/s	$0.2/M
47	Kimi K2.7 Code Moonshot AIAperto	41.8	—	42.0	39.9	39 t/s	$4/M
48	Qwen3 6 Max AlibabaAperto	41.1	—	—	40.0	85 t/s	$2/M
49	Deepseek V4 Flash 0420 High DeepSeekAperto	39.0	—	—	25.0	60 t/s	$1.1/M
50	Deepseek V3 2 Reasoning DeepSeekApertoreasoning	38.0	—	—	40.0	40 t/s	$1.1/M
51	Mimo V2 5 0424 XiaomiAperto	38.0	—	—	40.0	60 t/s	$3/M
52	Qwen3 6 27b AlibabaAperto	37.7	—	—	40.0	85 t/s	$2/M
53	Glm 5 1 Non Reasoning Z.aiApertoreasoning	36.3	—	—	40.0	40 t/s	$3/M
54	Gemma 4 31B GoogleAperto	35.0	1451	29.0	—	35 t/s	$0.3/M
55	Glm 5 2 Non Reasoning Z.aiApertoreasoning	34.8	—	—	40.0	40 t/s	$3/M
56	Glm 5 Non Reasoning Z.aiApertoreasoning	33.2	—	—	40.0	40 t/s	$3/M
57	Qwen3 5 397b A17b Non Reasoning AlibabaApertoreasoning	32.7	—	—	40.0	55 t/s	$2/M
58	Minimax M2 1 MiniMaxAperto	32.1	—	—	25.0	60 t/s	$2/M
59	Qwen3 6 35b A3b AlibabaAperto	32.1	—	—	40.0	85 t/s	$2/M
60	Deepseek V4 Pro 0424 Non Reasoning DeepSeekApertoreasoning	31.9	—	—	40.0	40 t/s	$1.1/M
61	Mimo V2 Flash Reasoning XiaomiApertoreasoning	31.9	—	—	25.0	40 t/s	$3/M
62	Qwen3 5 Omni Plus AlibabaAperto	31.3	—	—	40.0	85 t/s	$2/M
63	Qwen3 6 27b Non Reasoning AlibabaApertoreasoning	31.3	—	—	40.0	55 t/s	$2/M
64	Deepseek V3 1 Terminus Reasoning DeepSeekApertoreasoning	31.1	—	—	40.0	40 t/s	$1.1/M
65	GPT-oss-120B OpenAIAperto	30.0	—	—	37.0	240 t/s	$0.4/M
66	Qwen3 5 27b Non Reasoning AlibabaApertoreasoning	30.0	—	—	40.0	55 t/s	$2/M
67	Deepseek V4 Flash Non Reasoning DeepSeekApertoreasoning	29.3	—	—	25.0	40 t/s	$1.1/M
68	Glm 4 6 Reasoning Z.aiApertoreasoning	29.3	—	—	40.0	40 t/s	$3/M
69	Minimax M2 MiniMaxAperto	28.9	—	—	25.0	60 t/s	$2/M
70	Mimo V2 5 Pro Non Reasoning XiaomiApertoreasoning	28.4	—	—	40.0	40 t/s	$3/M
71	Qwen3 5 122b A10b Non Reasoning AlibabaApertoreasoning	28.2	—	—	40.0	55 t/s	$2/M
72	Glm 4 7 Non Reasoning Z.aiApertoreasoning	27.1	—	—	40.0	40 t/s	$3/M
73	Gemma 4 26b A4b GoogleAperto	26.1	—	—	40.0	120 t/s	$5/M
74	Deepseek V3 2 Reasoning 0925 DeepSeekApertoreasoning	25.9	—	—	40.0	40 t/s	$1.1/M
75	Qwen3 Max Thinking Preview AlibabaApertoreasoning	25.5	—	—	40.0	55 t/s	$2/M
76	Qwen3 6 35b A3b Non Reasoning AlibabaApertoreasoning	24.6	—	—	40.0	55 t/s	$2/M
77	Qwen3 5 35b A3b Non Reasoning AlibabaApertoreasoning	24.3	—	—	40.0	55 t/s	$2/M
78	Qwen3.5 397B-A17B AlibabaAperto	23.9	1442	34.0	11.5	71 t/s	$2/M
79	Gemma 4 31b Non Reasoning GoogleApertoreasoning	22.3	—	—	40.0	80 t/s	$5/M
80	Gemma 4 12b GoogleAperto	22.2	—	—	40.0	120 t/s	$5/M
81	Qwen3.5 122B-A10B AlibabaAperto	22.0	1417	32.0	—	126 t/s	$2/M
82	GPT-oss-20B OpenAIAperto	22.0	—	—	27.0	90 t/s	$0.2/M
83	Qwen3 5 9b AlibabaAperto	21.8	—	—	40.0	85 t/s	$2/M
84	Deepseek V3 2 0925 DeepSeekAperto	21.7	—	—	40.0	60 t/s	$1.1/M
85	Qwen3 Coder Next AlibabaAperto	21.3	—	—	40.0	85 t/s	$2/M
86	Deepseek V3 1 Reasoning DeepSeekApertoreasoning	21.0	—	—	40.0	40 t/s	$1.1/M
87	Qwen3 Vl 235b A22b Reasoning AlibabaApertoreasoning	20.9	—	—	40.0	55 t/s	$2/M
88	Qwen3 5 9b Non Reasoning AlibabaApertoreasoning	20.6	—	—	40.0	55 t/s	$2/M
89	Gemma 4 26b A4b Non Reasoning GoogleApertoreasoning	20.4	—	—	40.0	80 t/s	$5/M
90	Qwen3 5 4b AlibabaAperto	20.4	—	—	40.0	85 t/s	$2/M
91	Qwen3 235b A22b Instruct 2507 Reasoning AlibabaApertoreasoning	19.9	—	—	40.0	55 t/s	$2/M
92	Mistral Small 4 MistralAperto	19.7	—	—	40.0	100 t/s	$6/M
93	Qwen3 Max AlibabaAperto	19.6	1435	—	—	85 t/s	$2/M
94	Qwen3 Max Preview AlibabaAperto	19.4	—	—	40.0	85 t/s	$2/M
95	Qwen3 5 Omni Flash AlibabaAperto	19.2	—	—	25.0	85 t/s	$2/M
96	Nemotron 3 Ultra 550B NVIDIAAperto	19.0	1426	38.0	19.1	142 t/s	$2.75/M
97	Deepseek R1 0120 DeepSeekApertoreasoning	18.6	—	—	40.0	40 t/s	$1.1/M
98	Qwen3 235b A22b Instruct 2507 AlibabaAperto	18.4	—	—	40.0	85 t/s	$2/M
99	Qwen3 Coder 480b A35b Instruct AlibabaAperto	18.2	—	—	40.0	85 t/s	$2/M
100	Qwen3 Vl 32b Reasoning AlibabaApertoreasoning	18.1	—	—	40.0	55 t/s	$2/M

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

Model Name

Provider

Podium Score0.0

Context128K

Price (1M out)$0/M

Speed0 t/s

Architecture & Highlights

Model overview and workload profile.

Quickstart cURL

curl https://api.openrouter.ai/api/v1/chat/completions ...