📊 Live Benchmark

KI-Modelle Leaderboard 2026

Alle relevanten Large Language Models auf einen Blick — sortiert nach Quality Index, Geschwindigkeit, Latenz und Preis. Datenquelle: Artificial Analysis.

200 aktive Modelle aus 12 Vendor-Familien · Letzte Synchronisation: 11. Juli 2026

Wie liest du dieses Leaderboard?

Quality Index ist ein zusammengesetzter Wert von Artificial Analysis aus über zehn unabhängigen Benchmarks (MMLU-Pro, GPQA, HLE, LiveCodeBench, SciCode, AIME u. a.). Höher = besser. Speed misst Output-Tokens pro Sekunde im Median über alle Hosting-Provider. Latency ist die Time-to-First-Token in Sekunden — wichtig für Streaming-UIs. Preise sind US-Dollar pro 1 Million Tokens, separat für Input und Output. Sortiere nach deiner Priorität, filter nach Anbieter oder Preis-Bucket — und prüfe konkrete Kosten direkt im Token-Rechner oder zwei Modelle im Head-to-Head-Vergleich.

#	Modell	Vendor	Quality	Speed	Latency	Preis (USD/1M)
1	Anthropic: Claude Fable 5 anthropic/claude-fable-5	Anthropic	59,9	72,1 t/s	71,29 s	$10.00 in $50.00 out
2	OpenAI: GPT-5.6 Sol openai/gpt-5.6-sol	OpenAI	57,7	69,0 t/s	9,13 s	$5.00 in $30.00 out
3	Anthropic: Claude Opus 4.8 anthropic/claude-opus-4.8	Anthropic	55,7	57,1 t/s	30,42 s	$5.00 in $25.00 out
4	xAI: Grok 4.5 x-ai/grok-4.5	xAI	53,8	108,9 t/s	13,09 s	$2.00 in $6.00 out
5	Anthropic: Claude Opus 4.7 anthropic/claude-opus-4.7	Anthropic	53,5	48,5 t/s	17,78 s	$5.00 in $25.00 out
6	Google: Gemini 3.5 Flash google/gemini-3.5-flash	Google	50,2	236,7 t/s	11,02 s	$1.50 in $9.00 out
7	OpenAI: GPT-5.6 Terra openai/gpt-5.6-terra	OpenAI	49	116,5 t/s	1,65 s	$2.50 in $15.00 out
8	Google: Gemini 3.1 Pro Preview google/gemini-3.1-pro-preview	Google	46,5	131,2 t/s	19,75 s	$2.00 in $12.00 out
9	MiniMax: MiniMax M3 minimax/minimax-m3	MiniMax	44,4	106,3 t/s	1,40 s	$0.30 in $1.20 out
10	OpenAI: GPT-5.3-Codex openai/gpt-5.3-codex	OpenAI	44,3	83,3 t/s	81,50 s	$1.75 in $14.00 out
11	OpenAI: GPT-5.5 openai/gpt-5.5	OpenAI	43,5	60,7 t/s	1,75 s	$5.00 in $30.00 out
12	OpenAI: GPT-5.2 openai/gpt-5.2	OpenAI	42,2	75,9 t/s	75,51 s	$1.75 in $14.00 out
13	Anthropic: Claude Sonnet 5 anthropic/claude-sonnet-5	Anthropic	41,7	59,3 t/s	1,29 s	$2.00 in $10.00 out
14	Xiaomi: MiMo-V2-Pro xiaomi/mimo-v2-pro	Xiaomi	40,3	0 t/s	0 ms	$1.00 in $3.00 out
15	OpenAI: GPT-5.2-Codex openai/gpt-5.2-codex	OpenAI	40,1	129,1 t/s	1,15 s	$1.75 in $14.00 out
16	OpenAI: GPT-5.6 Luna openai/gpt-5.6-luna	OpenAI	38,1	217,9 t/s	1,45 s	$1.00 in $6.00 out
17	MiniMax: MiniMax M2.7 minimax/minimax-m2.7	MiniMax	38,1	49,7 t/s	1,42 s	$0.30 in $1.20 out
18	Anthropic: Claude Opus 4.6 anthropic/claude-opus-4.6	Anthropic	37,8	42,2 t/s	2,02 s	$5.00 in $25.00 out
19	NVIDIA: Nemotron 3 Ultra nvidia/nemotron-3-ultra-550b-a55b	NVIDIA	37,8	170,7 t/s	849 ms	$0.675 in $2.675 out
20	xAI: Grok 4.3 x-ai/grok-4.3	xAI	37,6	109,5 t/s	36,51 s	$1.25 in $2.50 out
21	OpenAI: GPT-5 Codex openai/gpt-5-codex	OpenAI	36,1	162,4 t/s	6,95 s	$1.25 in $10.00 out
22	Anthropic: Claude Sonnet 4.6 anthropic/claude-sonnet-4.6	Anthropic	35,9	46,5 t/s	1,05 s	$3.00 in $15.00 out
23	Xiaomi: MiMo-V2-Omni xiaomi/mimo-v2-omni	Xiaomi	35	0 t/s	0 ms	$0 in $0 out
24	OpenAI: GPT-5.1-Codex openai/gpt-5.1-codex	OpenAI	34,7	184,6 t/s	3,56 s	$1.25 in $10.00 out
25	OpenAI: GPT-5 openai/gpt-5	OpenAI	34,7	86,7 t/s	74,17 s	$1.25 in $10.00 out
26	Anthropic: Claude Opus 4.5 anthropic/claude-opus-4.5	Anthropic	34,7	53,0 t/s	1,30 s	$5.00 in $25.00 out
27	MiniMax: MiniMax M2.5 minimax/minimax-m2.5	MiniMax	33,7	75,9 t/s	1,40 s	$0.30 in $1.20 out
28	Tencent: Hy3 tencent/hy3	Tencent	33,6	196,0 t/s	1,74 s	$0.123 in $0.43 out
29	xAI: Grok 4 x-ai/grok-4	xAI	33,3	0 t/s	0 ms	$5.50 in $27.50 out
30	OpenAI: o3 Pro openai/o3-pro	OpenAI	32,5	24,2 t/s	72,57 s	$20.00 in $80.00 out
31	DeepSeek: DeepSeek V3.2 deepseek/deepseek-v3.2	DeepSeek	32	0 t/s	0 ms	$0.28 in $0.42 out
32	MiniMax: MiniMax M2.1 minimax/minimax-m2.1	MiniMax	31,4	77,2 t/s	1,18 s	$0.30 in $1.20 out
33	DeepSeek: DeepSeek V4 Pro deepseek/deepseek-v4-pro	DeepSeek	31,2	61,7 t/s	1,05 s	$0.435 in $0.87 out
34	Xiaomi: MiMo-V2-Flash xiaomi/mimo-v2-flash	Xiaomi	31,2	0 t/s	0 ms	$0.10 in $0.30 out
35	OpenAI: GPT-5 Mini openai/gpt-5-mini	OpenAI	30,9	103,4 t/s	14,60 s	$0.25 in $2.00 out
36	inclusionAI: Ring-2.6-1T inclusionai/ring-2.6-1t	Inclusionai	30,6	125,3 t/s	1,81 s	$0.30 in $2.50 out
37	OpenAI: GPT-5.1-Codex-Mini openai/gpt-5.1-codex-mini	OpenAI	30,6	211,6 t/s	3,33 s	$0.25 in $2.00 out
38	DeepSeek: DeepSeek V3.1 Terminus deepseek/deepseek-v3.1-terminus	DeepSeek	30,4	0 t/s	0 ms	$1.635 in $2.75 out
39	OpenAI: o3 openai/o3	OpenAI	30,4	107,1 t/s	6,62 s	$2.00 in $8.00 out
40	StepFun: Step 3.7 Flash stepfun/step-3.7-flash	Stepfun	30,3	401,5 t/s	733 ms	$0.20 in $1.15 out
41	OpenAI: GPT-5.4 Nano openai/gpt-5.4-nano	OpenAI	30,2	167,9 t/s	3,37 s	$0.20 in $1.25 out
42	Mistral: Mistral Medium 3.5 mistralai/mistral-medium-3-5	Mistral	29,9	92,1 t/s	761 ms	$1.50 in $7.50 out
43	Anthropic: Claude Sonnet 4.5 anthropic/claude-sonnet-4.5	Anthropic	29,3	45,8 t/s	1,48 s	$3.00 in $15.00 out
44	DeepSeek: DeepSeek V4 Flash deepseek/deepseek-v4-flash	DeepSeek	28,7	105 t/s	906 ms	$0.14 in $0.28 out
45	MiniMax: MiniMax M2 minimax/minimax-m2	MiniMax	28,3	76,5 t/s	1,22 s	$0.30 in $1.20 out
46	Anthropic: Claude Opus 4.1 anthropic/claude-opus-4.1	Anthropic	28,2	36,5 t/s	2,03 s	$15.00 in $75.00 out
47	Xiaomi: MiMo-V2.5-Pro xiaomi/mimo-v2.5-pro	Xiaomi	27,9	63,4 t/s	1,34 s	$0.661 in $1.722 out
48	OpenAI: GPT-5.4 openai/gpt-5.4	OpenAI	27,7	103,9 t/s	680 ms	$2.50 in $15.00 out
49	xAI: Grok 4 Fast x-ai/grok-4-fast	xAI	27,4	0 t/s	0 ms	$0.20 in $0.50 out
50	inclusionAI: Ling-2.6-1T inclusionai/ling-2.6-1t	Inclusionai	26,1	0 t/s	0 ms	$0.30 in $2.50 out
51	StepFun: Step 3.5 Flash stepfun/step-3.5-flash	Stepfun	26	195,5 t/s	889 ms	$0.10 in $0.30 out
52	Google: Gemini 2.5 Pro google/gemini-2.5-pro	Google	25,8	131,1 t/s	20,52 s	$1.25 in $10.00 out
53	OpenAI: o4 Mini openai/o4-mini	OpenAI	25,6	161,6 t/s	20,34 s	$1.10 in $4.40 out
54	Anthropic: Claude Opus 4 anthropic/claude-opus-4	Anthropic	25,5	0 t/s	0 ms	$15.00 in $75.00 out
55	Anthropic: Claude Sonnet 4 anthropic/claude-sonnet-4	Anthropic	25,5	0 t/s	0 ms	$3.00 in $15.00 out
56	NVIDIA: Nemotron 3 Super nvidia/nemotron-3-super-120b-a12b	NVIDIA	25,4	145,9 t/s	933 ms	$0.25 in $0.775 out
57	Inception: Mercury 2 inception/mercury-2	Inception	25,3	1042,5 t/s	4,10 s	$0.25 in $0.75 out
58	Google: Gemini 3.1 Flash Lite Preview google/gemini-3.1-flash-lite-preview	Google	25	308,8 t/s	5,36 s	$0.25 in $1.50 out
59	OpenAI: gpt-oss-120b openai/gpt-oss-120b	OpenAI	23,8	293,6 t/s	515 ms	$0.15 in $0.60 out
60	Anthropic: Claude Haiku 4.5 anthropic/claude-haiku-4.5	Anthropic	23,7	96,9 t/s	781 ms	$1.00 in $5.00 out
61	Anthropic: Claude 3.7 Sonnet anthropic/claude-3.7-sonnet	Anthropic	23,5	0 t/s	0 ms	$3.00 in $15.00 out
62	OpenAI: o1 openai/o1	OpenAI	23,4	138,5 t/s	16,68 s	$15.00 in $60.00 out
63	xAI: Grok 3 Mini x-ai/grok-3-mini	xAI	22,5	47,1 t/s	632 ms	$0.30 in $0.50 out
64	DeepSeek: DeepSeek V3.2 Speciale deepseek/deepseek-v3.2-speciale	DeepSeek	22,2	0 t/s	0 ms	$0 in $0 out
65	xAI: Grok 4.20 x-ai/grok-4.20	xAI	21,8	147,3 t/s	440 ms	$2.00 in $6.00 out
66	xAI: Grok Code Fast 1 x-ai/grok-code-fast-1	xAI	21,6	0 t/s	0 ms	$0 in $0 out
67	OpenAI: GPT-5.1 openai/gpt-5.1	OpenAI	20,4	97,9 t/s	774 ms	$1.25 in $10.00 out
68	Google: Gemini 2.5 Flash google/gemini-2.5-flash	Google	20,1	232,5 t/s	14,08 s	$0.30 in $2.50 out
69	DeepSeek: R1 deepseek/deepseek-r1	DeepSeek	20,1	0 t/s	0 ms	$1.35 in $4.20 out
70	OpenAI: GPT-5 Nano openai/gpt-5-nano	OpenAI	19,9	151,3 t/s	95,13 s	$0.05 in $0.40 out
71	OpenAI: GPT-4.1 openai/gpt-4.1	OpenAI	19,4	115,5 t/s	686 ms	$2.00 in $8.00 out
72	OpenAI: o3 Mini openai/o3-mini	OpenAI	19	221,3 t/s	4,84 s	$1.10 in $4.40 out
73	OpenAI: o1-pro openai/o1-pro	OpenAI	18,9	0 t/s	0 ms	$150.00 in $600.00 out
74	Perplexity: Sonar Reasoning Pro perplexity/sonar-reasoning-pro	Perplexity	17,8	0 t/s	0 ms	$0 in $0 out
75	xAI: Grok 4.1 Fast x-ai/grok-4.1-fast	xAI	16,9	0 t/s	0 ms	$0 in $0 out
76	OpenAI: GPT-5.4 Mini openai/gpt-5.4-mini	OpenAI	16,6	162,6 t/s	625 ms	$0.75 in $4.50 out
77	Prime Intellect: INTELLECT-3 prime-intellect/intellect-3	Prime-intellect	15,6	0 t/s	0 ms	$0 in $0 out
78	xAI: Grok 3 x-ai/grok-3	xAI	15,1	0 t/s	0 ms	$0 in $0 out
79	OpenAI: gpt-oss-20b openai/gpt-oss-20b	OpenAI	14,9	225,6 t/s	406 ms	$0.05 in $0.20 out
80	OpenAI: GPT-4.1 Mini openai/gpt-4.1-mini	OpenAI	14,8	73,9 t/s	606 ms	$0.40 in $1.60 out
81	Mistral: Mistral Medium 3.1 mistralai/mistral-medium-3.1	Mistral	14,7	84,0 t/s	504 ms	$0.40 in $2.00 out
82	Meta: Llama 4 Maverick meta-llama/llama-4-maverick	Meta	14,3	114,4 t/s	633 ms	$0.35 in $0.85 out
83	inclusionAI: Ling-2.6-flash inclusionai/ling-2.6-flash	Inclusionai	14,1	180,9 t/s	740 ms	$0.10 in $0.30 out
84	Upstage: Solar Pro 3 upstage/solar-pro-3	Upstage	14,1	0 t/s	0 ms	$0 in $0 out
85	Google: Gemini 2.5 Flash Lite Preview 09-2025 google/gemini-2.5-flash-lite-preview-09-2025	Google	13,1	0 t/s	0 ms	$0.10 in $0.40 out
86	Mistral: Mistral Medium 3 mistralai/mistral-medium-3	Mistral	12,5	46,7 t/s	535 ms	$0.40 in $2.00 out
87	Anthropic: Claude 3.5 Haiku anthropic/claude-3.5-haiku	Anthropic	12,3	0 t/s	0 ms	$0.80 in $4.00 out
88	Google: Gemini 2.5 Flash Lite google/gemini-2.5-flash-lite	Google	11,4	244,4 t/s	19,48 s	$0.10 in $0.40 out
89	OpenAI: GPT-4o openai/gpt-4o	OpenAI	11,2	199,7 t/s	598 ms	$2.50 in $10.00 out
90	DeepSeek: R1 Distill Qwen 32B deepseek/deepseek-r1-distill-qwen-32b	DeepSeek	11	0 t/s	0 ms	$0 in $0 out
91	Meta: Llama 4 Scout meta-llama/llama-4-scout	Meta	10	106,0 t/s	644 ms	$0.17 in $0.66 out
92	DeepSeek: R1 Distill Llama 70B deepseek/deepseek-r1-distill-llama-70b	DeepSeek	9,9	29,9 t/s	494 ms	$0.70 in $1.05 out
93	OpenAI: GPT-4.1 Nano openai/gpt-4.1-nano	OpenAI	9,6	135,4 t/s	526 ms	$0.10 in $0.40 out
94	OpenAI: GPT-4o (2024-08-06) openai/gpt-4o-2024-08-06	OpenAI	9,6	109,8 t/s	616 ms	$2.50 in $10.00 out
95	Perplexity: Sonar perplexity/sonar	Perplexity	9,5	0 t/s	0 ms	$0 in $0 out
96	Meta: Llama 3.3 70B Instruct meta-llama/llama-3.3-70b-instruct	Meta	9,4	90,2 t/s	620 ms	$0.58 in $0.71 out
97	Mistral: Devstral Small 1.1 mistralai/devstral-small	Mistral	9,3	0 t/s	0 ms	$0.10 in $0.30 out
98	Perplexity: Sonar Pro perplexity/sonar-pro	Perplexity	9,3	0 t/s	0 ms	$0 in $0 out
99	NVIDIA: Llama 3.1 Nemotron Ultra 253B v1 nvidia/llama-3.1-nemotron-ultra-253b-v1	NVIDIA	9,1	51,8 t/s	714 ms	$0.60 in $1.80 out
100	Baidu: ERNIE 4.5 300B A47B baidu/ernie-4.5-300b-a47b	Baidu	9	0 t/s	0 ms	$0.28 in $1.10 out
101	Google: Gemini 2.0 Flash Lite google/gemini-2.0-flash-lite-001	Google	8,8	0 t/s	0 ms	$0 in $0 out
102	NVIDIA: Nemotron Nano 9B V2 nvidia/nemotron-nano-9b-v2	NVIDIA	8,8	82,3 t/s	4,56 s	$0.04 in $0.16 out
103	OpenAI: GPT-4o (2024-05-13) openai/gpt-4o-2024-05-13	OpenAI	8,6	90,5 t/s	686 ms	$5.00 in $15.00 out
104	Mistral: Pixtral Large 2411 mistralai/pixtral-large-2411	Mistral	8,1	0 t/s	0 ms	$2.00 in $6.00 out
105	OpenAI: GPT-4 Turbo openai/gpt-4-turbo	OpenAI	7,9	33,6 t/s	1,32 s	$10.00 in $30.00 out
106	Cohere: Command A cohere/command-a	Cohere	7,7	56,4 t/s	444 ms	$2.50 in $10.00 out
107	NVIDIA: Llama 3.1 Nemotron 70B Instruct nvidia/llama-3.1-nemotron-70b-instruct	NVIDIA	7,6	301 t/s	2,49 s	$1.20 in $1.20 out
108	Meta: Llama 3.1 8B Instruct meta-llama/llama-3.1-8b-instruct	Meta	7,6	156,8 t/s	510 ms	$0.075 in $0.095 out
109	NVIDIA: Nemotron 3 Nano 30B A3B nvidia/nemotron-3-nano-30b-a3b	NVIDIA	7,4	118,1 t/s	306 ms	$0.05 in $0.20 out
110	Mistral Large 2407 mistralai/mistral-large-2407	Mistral	7,3	0 t/s	0 ms	$2.00 in $6.00 out
111	OpenAI: GPT-4 openai/gpt-4	OpenAI	7	31,1 t/s	1,05 s	$30.00 in $60.00 out
112	OpenAI: GPT-4o-mini openai/gpt-4o-mini	OpenAI	6,9	70,6 t/s	667 ms	$0.15 in $0.60 out
113	Meta: Llama 3.1 70B Instruct meta-llama/llama-3.1-70b-instruct	Meta	6,8	34,1 t/s	606 ms	$0.56 in $0.56 out
114	Mistral: Saba mistralai/mistral-saba	Mistral	6,4	0 t/s	0 ms	$0 in $0 out
115	NVIDIA: Nemotron Nano 12B 2 VL nvidia/nemotron-nano-12b-v2-vl	NVIDIA	4,6	215,3 t/s	744 ms	$0.20 in $0.60 out
116	Mistral Large mistralai/mistral-large	Mistral	4,4	0 t/s	0 ms	$4.00 in $12.00 out
117	Meta: Llama 3.2 3B Instruct meta-llama/llama-3.2-3b-instruct	Meta	4,2	52,7 t/s	609 ms	$0.15 in $0.15 out
118	Anthropic: Claude 3 Haiku anthropic/claude-3-haiku	Anthropic	3,9	0 t/s	0 ms	$0.25 in $1.25 out
119	Meta: Llama 3 70B Instruct meta-llama/llama-3-70b-instruct	Meta	3,5	45,6 t/s	670 ms	$0.65 in $2.75 out
120	Meta: Llama 3.2 11B Vision Instruct meta-llama/llama-3.2-11b-vision-instruct	Meta	3,3	67,7 t/s	544 ms	$0.345 in $0.345 out
121	Mistral: Mixtral 8x7B Instruct mistralai/mixtral-8x7b-instruct	Mistral	2,4	0 t/s	0 ms	$0.45 in $0.70 out
122	Meta: Llama 3 8B Instruct meta-llama/llama-3-8b-instruct	Meta	1,2	78,0 t/s	492 ms	$0.045 in $0.145 out
123	Meta: Llama 3.2 1B Instruct meta-llama/llama-3.2-1b-instruct	Meta	1,1	86,6 t/s	592 ms	$0.05 in $0.05 out
124	OpenAI: GPT-5.6 Luna Pro openai/gpt-5.6-luna-pro	OpenAI	—	— t/s	—	— in — out
125	OpenAI: GPT-5.6 Terra Pro openai/gpt-5.6-terra-pro	OpenAI	—	— t/s	—	— in — out
126	Elephant openrouter/elephant-alpha	Openrouter	—	— t/s	—	— in — out
127	Qwen: Qwen3.7 Max qwen/qwen3.7-max	Alibaba / Qwen	—	— t/s	—	— in — out
128	OpenAI: GPT-5.6 Sol Pro openai/gpt-5.6-sol-pro	OpenAI	—	— t/s	—	— in — out
129	xAI: Grok Build 0.1 x-ai/grok-build-0.1	xAI	—	— t/s	—	— in — out
130	Anthropic: Claude Opus 4.7 (Fast) anthropic/claude-opus-4.7-fast	Anthropic	—	— t/s	—	— in — out
131	inclusionAI: Ring-2.6-1T (free) inclusionai/ring-2.6-1t:free	Inclusionai	—	— t/s	—	— in — out
132	Poolside: Laguna M.1 poolside/laguna-m.1	Poolside	—	— t/s	—	— in — out
133	Anthropic Claude Haiku Latest ~anthropic/claude-haiku-latest	~anthropic	—	— t/s	—	— in — out
134	Nex AGI: Nex-N2-Pro (free) nex-agi/nex-n2-pro:free	Nex-agi	—	— t/s	—	— in — out
135	Baidu Qianfan: CoBuddy (free) baidu/cobuddy:free	Baidu	—	— t/s	—	— in — out
136	DeepSeek: DeepSeek V4 Flash (free) deepseek/deepseek-v4-flash:free	DeepSeek	—	— t/s	—	— in — out
137	Nex AGI: Nex-N2-Mini nex-agi/nex-n2-mini	Nex-agi	—	— t/s	—	— in — out
138	Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) google/gemini-3.1-flash-lite-image	Google	—	— t/s	—	— in — out
139	xAI: Grok Latest ~x-ai/grok-latest	~x-ai	—	— t/s	—	— in — out
140	AionLabs: Aion-3.0-Mini aion-labs/aion-3.0-mini	Aion-labs	—	— t/s	—	— in — out
141	AionLabs: Aion-3.0 aion-labs/aion-3.0	Aion-labs	—	— t/s	—	— in — out
142	OpenAI GPT Mini Latest ~openai/gpt-mini-latest	~openai	—	— t/s	—	— in — out
143	Google Gemini Pro Latest ~google/gemini-pro-latest	~google	—	— t/s	—	— in — out
144	Tencent: Hy3 (free) tencent/hy3:free	Tencent	—	— t/s	—	— in — out
145	Poolside: Laguna XS 2.1 (free) poolside/laguna-xs-2.1:free	Poolside	—	— t/s	—	— in — out
146	Poolside: Laguna XS 2.1 poolside/laguna-xs-2.1	Poolside	—	— t/s	—	— in — out
147	inclusionAI: Ling-2.6-flash (free) inclusionai/ling-2.6-flash:free	Inclusionai	—	— t/s	—	— in — out
148	Owl Alpha openrouter/owl-alpha	Openrouter	—	— t/s	—	— in — out
149	inclusionAI: Ling-2.6-1T (free) inclusionai/ling-2.6-1t:free	Inclusionai	—	— t/s	—	— in — out
150	Tencent: Hy3 preview (free) tencent/hy3-preview:free	Tencent	—	— t/s	—	— in — out
151	MoonshotAI Kimi Latest ~moonshotai/kimi-latest	~moonshotai	—	— t/s	—	— in — out
152	OpenAI: GPT-5.4 Image 2 openai/gpt-5.4-image-2	OpenAI	—	— t/s	—	— in — out
153	Sakana: Fugu Ultra sakana/fugu-ultra	Sakana	—	— t/s	—	— in — out
154	Google: Nano Banana 2 (Gemini 3.1 Flash Image) google/gemini-3.1-flash-image	Google	—	— t/s	—	— in — out
155	Google: Nano Banana Pro (Gemini 3 Pro Image) google/gemini-3-pro-image	Google	—	— t/s	—	— in — out
156	Cohere: North Mini Code (free) cohere/north-mini-code:free	Cohere	—	— t/s	—	— in — out
157	Z.ai: GLM 5.2 z-ai/glm-5.2	Z-ai	—	— t/s	—	— in — out
158	OpenRouter: Fusion openrouter/fusion	Openrouter	—	— t/s	—	— in — out
159	MoonshotAI: Kimi K2.7 Code moonshotai/kimi-k2.7-code	Moonshot AI	—	— t/s	—	— in — out
160	Anthropic: Claude Fable Latest ~anthropic/claude-fable-latest	~anthropic	—	— t/s	—	— in — out
161	Baidu: Qianfan-OCR-Fast baidu/qianfan-ocr-fast	Baidu	—	— t/s	—	— in — out
162	Poolside: Laguna XS.2 (free) poolside/laguna-xs.2:free	Poolside	—	— t/s	—	— in — out
163	Poolside: Laguna XS.2 poolside/laguna-xs.2	Poolside	—	— t/s	—	— in — out
164	Baidu: Qianfan-OCR-Fast (free) baidu/qianfan-ocr-fast:free	Baidu	—	— t/s	—	— in — out
165	Nex AGI: Nex-N2-Pro nex-agi/nex-n2-pro	Nex-agi	—	— t/s	—	— in — out
166	Arcee AI: Trinity Large Thinking (free) arcee-ai/trinity-large-thinking:free	Arcee-ai	—	— t/s	—	— in — out
167	NVIDIA: Nemotron 3.5 Content Safety (free) nvidia/nemotron-3.5-content-safety:free	NVIDIA	—	— t/s	—	— in — out
168	NVIDIA: Nemotron 3 Ultra (free) nvidia/nemotron-3-ultra-550b-a55b:free	NVIDIA	—	— t/s	—	— in — out
169	Qwen: Qwen3.7 Plus qwen/qwen3.7-plus	Alibaba / Qwen	—	— t/s	—	— in — out
170	Anthropic: Claude Opus 4.8 (Fast) anthropic/claude-opus-4.8-fast	Anthropic	—	— t/s	—	— in — out
171	Perceptron: Perceptron Mk1 perceptron/perceptron-mk1	Perceptron	—	— t/s	—	— in — out
172	Qwen: Qwen3.6 Plus (free) qwen/qwen3.6-plus:free	Alibaba / Qwen	—	— t/s	—	— in — out
173	Anthropic: Claude Opus 4.6 (Fast) anthropic/claude-opus-4.6-fast	Anthropic	—	— t/s	—	— in — out
174	MoonshotAI: Kimi K2.6 (free) moonshotai/kimi-k2.6:free	Moonshot AI	—	— t/s	—	— in — out
175	Google: Gemini 3.1 Flash Lite google/gemini-3.1-flash-lite	Google	—	— t/s	—	— in — out
176	Poolside: Laguna M.1 (free) poolside/laguna-m.1:free	Poolside	—	— t/s	—	— in — out
177	OpenAI: GPT Chat Latest openai/gpt-chat-latest	OpenAI	—	— t/s	—	— in — out
178	IBM: Granite 4.1 8B ibm-granite/granite-4.1-8b	Ibm-granite	—	— t/s	—	— in — out
179	NVIDIA: Nemotron 3 Nano Omni (free) nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free	NVIDIA	—	— t/s	—	— in — out
180	OpenAI GPT Latest ~openai/gpt-latest	~openai	—	— t/s	—	— in — out
181	Google Gemini Flash Latest ~google/gemini-flash-latest	~google	—	— t/s	—	— in — out
182	Tencent: Hy3 preview tencent/hy3-preview	Tencent	—	— t/s	—	— in — out
183	Qwen: Qwen3.5 Plus 2026-04-20 qwen/qwen3.5-plus-20260420	Alibaba / Qwen	—	— t/s	—	— in — out
184	MiniMax: MiniMax M2.5 (free) minimax/minimax-m2.5:free	MiniMax	—	— t/s	—	— in — out
185	Arcee AI: Trinity Large Preview arcee-ai/trinity-large-preview	Arcee-ai	—	— t/s	—	— in — out
186	Anthropic Claude Sonnet Latest ~anthropic/claude-sonnet-latest	~anthropic	—	— t/s	—	— in — out
187	AllenAI: Olmo 3.1 32B Instruct allenai/olmo-3.1-32b-instruct	Allenai	—	— t/s	—	— in — out
188	Qwen: Qwen3.6 Flash qwen/qwen3.6-flash	Alibaba / Qwen	—	— t/s	—	— in — out
189	Qwen: Qwen3.6 35B A3B qwen/qwen3.6-35b-a3b	Alibaba / Qwen	—	— t/s	—	— in — out
190	Xiaomi: MiMo-V2.5 xiaomi/mimo-v2.5	Xiaomi	—	— t/s	—	— in — out
191	Qwen: Qwen3.6 Max Preview qwen/qwen3.6-max-preview	Alibaba / Qwen	—	— t/s	—	— in — out
192	Qwen: Qwen3.6 27B qwen/qwen3.6-27b	Alibaba / Qwen	—	— t/s	—	— in — out
193	Google: Gemma 4 31B (free) google/gemma-4-31b-it:free	Google	—	— t/s	—	— in — out
194	Google: Gemma 4 26B A4B (free) google/gemma-4-26b-a4b-it:free	Google	—	— t/s	—	— in — out
195	Anthropic: Claude Opus Latest ~anthropic/claude-opus-latest	~anthropic	—	— t/s	—	— in — out
196	Google: Gemma 4 26B A4B google/gemma-4-26b-a4b-it	Google	—	— t/s	—	— in — out
197	Pareto Code Router openrouter/pareto-code	Openrouter	—	— t/s	—	— in — out
198	MoonshotAI: Kimi K2.6 moonshotai/kimi-k2.6	Moonshot AI	—	— t/s	—	— in — out
199	Z.ai: GLM 5.1 z-ai/glm-5.1	Z-ai	—	— t/s	—	— in — out
200	Meituan: LongCat Flash Chat meituan/longcat-flash-chat	Meituan	—	— t/s	—	— in — out

🧮 Tool

Pricing-Calculator

Schätze deine monatlichen API-Kosten — gib dein erwartetes Token-Volumen ein, wir rechnen für die Top-15 Modelle.

Input Tokens (Mio./Monat) Output Tokens (Mio./Monat)

#	Modell	Vendor	Quality	Kosten/Monat (USD)

📈 Tool

Quality vs. Preis (Pareto-Chart)

Wo liegt der beste Tradeoff? Modelle oben-links sind die Pareto-Optima: hohe Quality, niedriger Preis.

Pareto-Optimum Andere Modelle

❓ Häufige Fragen zum KI-Modelle-Leaderboard

Woher stammen die Leaderboard-Daten?

Quality Index, Speed (Output-Tokens/s), Latenz (Time-to-First-Token) und Preise (USD pro 1 Mio. Tokens) kommen direkt von Artificial Analysis. Synchronisation täglich um 04:00 UTC. Wir speichern keine eigenen Benchmark-Werte und führen keine eigene Bewertung durch.

Was misst der Quality Index genau?

Der Artificial-Analysis-Intelligence-Index ist ein zusammengesetzter Score aus über zehn unabhängigen Benchmarks: MMLU-Pro (Wissen), GPQA & HLE (Reasoning), LiveCodeBench & SciCode (Coding), AIME (Mathematik), IFBench (Instruction-Following), LCR (Long-Context-Recall) und τ² (Tool-Use). Skala 0–100 — höher = besser.

Warum sind manche bekannte Modelle nicht im Leaderboard?

Wir zeigen nur Modelle, die in der Registry als is_active=true markiert sind und für die Artificial Analysis vollständige Benchmark-Daten liefert. Reine Bild- oder Audio-Modelle, deprecatete Versionen sowie Closed-Beta-Modelle ohne öffentliche API erscheinen nicht.

Wie nutze ich Sortierung und Filter sinnvoll?

Sortiere nach Quality für maximale Genauigkeit, nach Speed für hohen Durchsatz, nach Latency für reaktive Streaming-UIs und nach Preis für kostensensitive Workloads. Kombiniere die Vendor- und Preis-Filter, um z. B. „nur OpenAI unter 5 USD/Mio. Output“ zu finden.

Wie berechne ich konkrete Token-Kosten?

Im Token-Kostenrechner kannst du Input- und Output-Volumen einsetzen und die monatlichen Kosten für jedes Modell live durchspielen. Für direkten Head-to-Head-Vergleich zweier Modelle nutze die Vergleichs-Seiten.