Naar inhoud

Benchmarks

Leaderboard

All active models ranked by P50 latency — the median response time for a standard 500-token output, measured from EU (Amsterdam). Green < 500 ms, yellow 500–1000 ms, red > 1000 ms.

Filter:
#
1pplC22
2Mistral-7B-Instruct-v0.3115
3Mistral-Nemo-Instruct-2407117
4Mistral-Small-3.2-24B-Instruct-2506120
5Qwen2.5-VL-72B-Instruct125
6Meta-Llama-3_3-70B-Instruct127
7Llama-3.1-8B-Instruct130
8Mistral Voxtral Small 24BA134
9Llama 4 ScoutA165
10Llama 4 MaverickA179
11Nous Hermes 3 70BA180
12NVIDIA Nemotron Super 49B v1.5A184
13Qwen 2.5 VL 72B InstructA192
14Qwen3.5-397B-A17B210
15gpt-oss-20bC257
16gpt-4.1-nanoC334
17gpt-oss-120bC403
18gpt-5-chat-latestC413
19Qwen3-32B425
20Gemini 2.5 Flash-LiteB454
21Qwen3.5-9B460
22o3-miniC466
23gpt-4o-miniC496
24gpt-4oC500
25Qwen3-Coder-30B-A3B-Instruct526
26gpt-4.1-miniC546
27gpt-5.4-miniA560
28gpt-5.1-chat-latestC573
29o4-miniC577
30Claude Haiku 4.5A591
31gpt-5.4-nanoC632
32o3C673
33Llama 3.3 70B InstructA750
34gpt-5.4A756
35gpt-5.2-chat-latestC793
36gpt-5-nanoC833
37gpt-5.3-chat-latestC875
38DeepSeek v3.2A919
39Google Lyria 3 Pro PreviewA942
40gpt-5.2B942
41Claude Opus 4.6B943
42Gemini 3.1 Flash Lite957
43Claude Opus 4.8A959
44gpt-5C965
45MiniMax M2.5A977
46gpt-5-miniC999
47Cohere Command-AA1035
48Qwen 3.7 MaxA1038
49gpt-4o-2024-05-13C1049
50Claude Sonnet 4.6A1064
51gpt-4.1-2025-04-141072
52gpt-4.1B1081
53gpt-5.5C1095
54gpt-5.1B1145
55Gemini 2.5 FlashA1258
56gpt-4o-2024-11-20C1326
57gpt-3.5-turbo-11061328
58Qwen 3.6 PlusA1340
59Gemini Flash-Lite LatestC1366
60DeepSeek v4 ProA1389
61Claude Sonnet 4.5B1483
62Claude Opus 4.7B1574
63Gemini 2.5 ProA1709
64Claude Opus 4.5B1711
65Nano Banana1808
66Nano Banana 21887
67Claude Opus 4.1C1932
68gpt-3.5-turboC1995
69gpt-3.5-turbo-16k2006
70gpt-4o-2024-08-06C2016
71gpt-4.1-nano-2025-04-142051
72Claude Opus 4C2093
73gpt-3.5-turbo-01252331
74Gemini Robotics-ER 1.6 Preview2764
75Gemini 3 Flash PreviewC2780
76gpt-4o-search-previewC2930
77gpt-4o-mini-search-previewC3388
78gpt-5-search-apiC3559
79gpt-4.1-mini-2025-04-143561
80Gemini 3.5 FlashA3938
81gpt-4o-mini-2024-07-18C3960
82Gemini Flash LatestB4051
83gpt-4o-mini-search-preview-2025-03-114627
84gpt-4o-search-preview-2025-03-114883
85gpt-5-search-api-2025-10-145351
86Claude Sonnet 4C5563
87gpt-4-06135810
88Gemini 3.1 Pro Preview Custom ToolsC6069
89Gemini Pro LatestC6574
90Gemini 3.1 Pro PreviewC6937
91gpt-4-turbo-2024-04-09C7386
92gpt-4C7408
93Nano Banana Pro8045
94gpt-4-turboC9151
95Lyria 3 Clip Preview9402
96Gemma 4 31B ITC11240
97Gemma 4 26B A4B ITC12943
98Lyria 3 Pro Preview21413

98 of 98 models · click column headers to sort

Fast (< 500 ms)
Medium (500–1000 ms)
Slow (> 1000 ms)
Updated every 6 hours · P50 = median latency · P95 = tail latency