Speed Champions Leaderboard

Top 20 models ranked by output speed (tok/s)

Rank	Model	Creator	Output Speed (tok/s)	Price/1M	Speed
🥇	Mercury 2 Inception	Inception	859 tok/s	$0.38	859 tok/s
🥈	Granite 3.3 8B (Non-reasoning) IBM	IBM	509 tok/s	$0.09	509 tok/s
🥉	Granite 4.0 H Small IBM	IBM	499 tok/s	$0.11	499 tok/s
4	NVIDIA Nemotron 3 Super 120B A12B (Reasoning) NVIDIA	NVIDIA	429 tok/s	$0.41	429 tok/s
5	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) Google	Google	396 tok/s	$0.17	396 tok/s
6	Nova Micro Amazon	Amazon	336 tok/s	$0.06	336 tok/s
7	Gemini 2.5 Flash-Lite (Non-reasoning) Google	Google	322 tok/s	$0.17	322 tok/s
8	Sarvam 30B (Reasoning) Sarvam	Sarvam	311 tok/s	$0.00	311 tok/s
9	gpt-oss-20B (low) OpenAI	OpenAI	306 tok/s	$0.09	306 tok/s
10	gpt-oss-20B (high) OpenAI	OpenAI	304 tok/s	$0.09	304 tok/s
11	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) Google	Google	281 tok/s	$0.17	281 tok/s
12	gpt-oss-120B (high) OpenAI	OpenAI	277 tok/s	$0.26	277 tok/s
13	gpt-oss-120B (low) OpenAI	OpenAI	273 tok/s	$0.26	273 tok/s
14	Gemini 2.5 Flash-Lite (Reasoning) Google	Google	271 tok/s	$0.17	271 tok/s
15	GPT-5.4 mini (medium) OpenAI	OpenAI	270 tok/s	$1.69	270 tok/s
16	Ministral 3 3B Mistral	Mistral	261 tok/s	$0.10	261 tok/s
17	Devstral Small (Jul '25) Mistral	Mistral	261 tok/s	$0.15	261 tok/s
18	Gemini 3.1 Flash-Lite Preview Google	Google	260 tok/s	$0.56	260 tok/s
19	GPT-5.4 mini (xhigh) OpenAI	OpenAI	249 tok/s	$1.69	249 tok/s
20	LFM2 24B A2B Liquid AI	Liquid AI	228 tok/s	$0.05	228 tok/s