Output Speed (tokens/s)

Name: Output Speed (tokens/s)
Creator: aib vote
License: https://www.gnu.org/licenses/gpl-3.0.html

About This Benchmark

Number of output tokens generated per second via API. Higher values mean faster responses. Median measured by Artificial Analysis.

Source: Artificial Analysis

Rank	Model
#1	Google Gemini 3.1 Flash Lite	335 tok/s
#2	Amazon Nova 2 Lite	216 tok/s
#3	Google Gemini 2.5 Flash	213 tok/s
#4	OpenAI GPT-5.4 Nano	197 tok/s
#5	Google Gemini 3 Flash	177 tok/s
#6	OpenAI GPT-5.4 Mini	172 tok/s
#7	OpenAI GPT-5.4	155 tok/s
#8	Mistral AI Mistral Small 4	151 tok/s
#9	Meta Llama 4 Scout	136 tok/s
#10	OpenAI GPT-5 Nano	135 tok/s
#11	Google Gemini 3.1 Pro	135 tok/s
#12	Grok Grok 4.1 Fast	134 tok/s
#13	Google Gemini 2.5 Pro	131 tok/s
#14	Arcee AI Trinity Large Thinking	129 tok/s
#15	Google Nano Banana	122 tok/s
#16	Meta Llama 4 Maverick	119 tok/s
#17	Meituan Longcat Flash Chat	115 tok/s
#18	Grok Grok 4.1 Fast (Reasoning)	114 tok/s
#19	Grok Grok 4.20 (Reasoning)	113 tok/s
#20	Grok Grok 4.20	107 tok/s
#21	Google Gemini 2.5 Flash Lite	105 tok/s
#22	MiniMax MiniMax M2.5	104 tok/s
#23	OpenAI GPT-4.1	103 tok/s
#24	Anthropic Claude Haiku 4.5	99 tok/s
#25	OpenAI GPT OSS 120B	86 tok/s
#26	Google Nano Banana 2	82 tok/s
#27	NVIDIA Nemotron 3 Super	80 tok/s
#28	OpenAI GPT-5 Mini	79 tok/s
#29	OpenAI GPT-5	77 tok/s
#30	Xiaomi MiMo-V2-Pro	67 tok/s
#31	Anthropic Claude Sonnet 4.6	59 tok/s
#32	Anthropic Claude Opus 4.7	56 tok/s
#33	OpenAI GPT-4o Mini Transcribe	53 tok/s
#34	Alibaba Qwen3.6 Plus	53 tok/s
#35	Z.ai GLM-5	52 tok/s
#36	Alibaba Qwen3.5 397B A17B	52 tok/s
#37	Anthropic Claude Opus 4.5	51 tok/s
#38	MiniMax MiniMax M2.7	47 tok/s
#39	Z.ai GLM-5.1	47 tok/s
#40	DeepSeek DeepSeek V3.2	47 tok/s
#41	Anthropic Claude Opus 4.6	45 tok/s
#42	Anthropic Claude Sonnet 4	45 tok/s
#43	Anthropic Claude Sonnet 4.5	44 tok/s
#44	Anthropic Claude Opus 4	34 tok/s
#45	Anthropic Claude Opus 4.1	34 tok/s
#46	DeepSeek DeepSeek V4 Flash	33 tok/s
#47	Moonshot AI Kimi K2.5	33 tok/s
#48	DeepSeek DeepSeek V4 Pro	32 tok/s
#49	OpenAI GPT-4o Transcribe	31 tok/s
#50	Google Nano Banana Pro	28 tok/s
#51	Moonshot AI Kimi K2.6	26 tok/s
#52	Baidu ERNIE 4.5 300B A47B	24 tok/s
#53	Google Gemma 4 31B	14 tok/s
#54	OpenAI GPT-5.4 Pro	6 tok/s