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:
#
1FLUX.1 Kontext [max] — Multi-Image FusionA0
2FLUX.1 Kontext [pro] — Multi-Image FusionB0
3NVIDIA Nemotron Super 49B v1.5A26
4Qwen3-Coder-30B-A3B-InstructB92
5SDXL 1.0C100
6Llama 4 ScoutA110
7Mistral-Small-3.2-24B-Instruct-2506B120
8Mistral-Nemo-Instruct-2407C124
9Meta-Llama-3_3-70B-InstructB133
10Mistral-7B-Instruct-v0.3C164
11Qwen2.5-VL-72B-InstructB174
12Mistral Voxtral Small 24BA181
13Llama 4 MaverickA198
14Qwen 2.5 VL 72B InstructA200
15Nous Hermes 3 70BA212
16gpt-oss-20bC240
17Qwen3.5-397B-A17BA297
18Llama 3.3 70B InstructA326
19Cohere Command-AA343
20Qwen3.5-9BB492
21gpt-5.4-nanoC504
22gpt-4o-miniC523
23MiniMax M2.5A525
24gpt-4.1B526
25FLUX.1 SchnellB559
26gpt-4oC588
27Gemini 2.5 FlashA596
28Gemini 2.5 Flash-LiteB601
29GLM-4.7A615
30gpt-oss-120bC635
31gpt-5.4-miniA690
32GLM ImageB714
33gpt-4.1-nano-2025-04-14C721
34DeepSeek v4 ProA753
35gpt-5.1B795
36Qwen3-32BB800
37o3C805
38Gemini 3.1 Flash LiteB812
39o4-miniC826
40Qwen 3.6 PlusA857
41gpt-5-miniC875
42gpt-5-nanoC915
43gpt-4.1-nanoC929
44gpt-5.4A977
45gpt-4o-2024-11-20C1003
46Gemini Flash-Lite LatestC1008
47gpt-5.5C1029
48gpt-5.4-mini-2026-03-17A1046
49GLM-4.6V (vision)A1057
50Qwen 3.7 MaxA1069
51gpt-5C1072
52gpt-5.2-chat-latestC1076
53gpt-3.5-turboC1078
54gpt-4o-mini-2024-07-18C1097
55Claude Opus 4.8A1121
56Claude Haiku 4.5A1150
57Claude Sonnet 4.6A1169
58gpt-4o-2024-08-06C1232
59gpt-3.5-turbo-1106C1249
60gpt-4o-mini-search-previewC1276
61gpt-5.1-2025-11-13B1317
62Claude Sonnet 4.5B1326
63Claude Opus 4.7B1345
64o3-miniC1354
65gpt-5.4-2026-03-05B1375
66gpt-5.2B1397
67o3-2025-04-16B1425
68gpt-4.1-2025-04-14C1445
69gpt-5.4-nano-2026-03-17A1461
70Claude Opus 5A1485
71gpt-3.5-turbo-0125C1486
72gpt-5.2-2025-12-11B1507
73gpt-4-0613C1524
74gpt-5-search-api-2025-10-14B1551
75DeepSeek v3.2A1652
76gpt-4o-2024-05-13C1695
77gpt-3.5-turbo-16kC1697
78Claude Opus 4.5B1712
79gpt-5.5-2026-04-23A1728
80o3-mini-2025-01-31C1738
81GLM-4.5V (vision)A1792
82Gemini 2.5 ProA1800
83GLM-5A1821
84gpt-5.3-chat-latestC1858
85Claude Opus 4.1C1970
86gpt-4o-search-previewC2032
87o4-mini-2025-04-16B2161
88gpt-4.1-mini-2025-04-14C2219
89GLM-4.6A2232
90gpt-5-mini-2025-08-07B2427
91o1C2530
92Claude Fable 5A2833
93Nano BananaB2873
94o1-2024-12-17C2940
95Gemini 3 Flash PreviewC2946
96gpt-5-nano-2025-08-07B2962
97Gemini 3.5 FlashA2990
98gpt-4C3002
99Claude Opus 4.6B3040
100GLM-5.1A3394
101Gemini Flash LatestB3571
102GLM-4.5 AirB3653
103gpt-5-search-apiC3713
104GLM-5.2A4038
105Gemini Robotics-ER 1.6 PreviewB4190
106Nano Banana 2B4330
107Gemini Pro LatestC4359
108gpt-5-2025-08-07B4774
109gpt-4-turboC4835
110Gemini 3.1 Pro PreviewC4955
111GLM-4.5A6228
112gpt-4-turbo-2024-04-09C6778
113Gemini 3.1 Pro Preview Custom ToolsC7298
114CogView-4B7569
115GLM-5 TurboB7609
116gpt-4.1-miniC9630
117Nano Banana ProA11201

117 of 117 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