Artificial Analysis Output Speed leaderboard: who is #1

Name: Artificial Analysis Output Speed rankings
Creator: Artificial Analysis
License: https://creativecommons.org/licenses/by/4.0/

Artificial Analysis · Output speed, higher is better · 24 models · published Sep 24, 2026

Top 24 of 24 models, best variant per model as published

#	Model	Score
1	Gemini 3.8 Flashas “Gemini 3.8 Flash (high)”	291 tok/s
2	DeepSeek V4.1 Flashas “DeepSeek V4.1 Flash (Reasoning, Max Effort)”	232 tok/s
3	Muse Spark 1.3as “Muse Spark 1.3 (xhigh)”	228 tok/s
4	DeepSeek V4 Flash Visionas “DeepSeek V4 Flash Vision (Reasoning, Max Effort)”	218 tok/s
5	GPT-5.3 Codexas “GPT-5.3 Codex (xhigh)”	145 tok/s
6	GPT-6 Lunaas “GPT-6 Luna (high)”	134 tok/s
7	GPT-6 Solas “GPT-6 Sol (medium)”	114 tok/s
8	Claude Opus 5.5as “Claude Opus 5.5 (Adaptive Reasoning, Xhigh Effort, Default Fallback)”	92 tok/s
9	Step 5 PreviewStepFun	85 tok/s
10	GPT-5.6 Terraas “GPT-5.6 Terra (max)”	83 tok/s
11	Claude Sonnet 5as “Claude Sonnet 5 (Adaptive Reasoning, Max Effort)”	79 tok/s
12	DeepSeek V4 Pro 0813as “DeepSeek V4 Pro 0813 (Reasoning, Max Effort)”	66 tok/s
13	Claude Fable 5.1as “Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)”	66 tok/s
14	Grok 4.6as “Grok 4.6 (high)”	61 tok/s
15	GLM-5.3as “GLM-5.3 (max)”	57 tok/s
16	Qwen3.8-Flash-NextAlibaba	54 tok/s
17	GPT-6 Astraas “GPT-6 Astra (max)”	52 tok/s
18	Grok 4.7as “Grok 4.7 (high)”	52 tok/s
19	MiMo-V2.6-ProXiaomi	49 tok/s
20	GLM 5.3 FlashZ.ai (Zhipu)	46 tok/s
21	Qwen3.8 27Bas “Qwen3.8 27B (xhigh)”	43 tok/s
22	Qwen3.8 Max (0902)Alibaba	39 tok/s
23	Kimi K3as “Kimi K3 (low)”	38 tok/s
24	Qwen3.8 2.4T A95BAlibaba	38 tok/s