AI Models Leaderboard 2026

Compare the latest published AI model snapshot across pricing, context windows, latency, and benchmark performance.

Model	Intelligence	MMLU Pro	Context	GPQA	TTFT (s)	Speed (T/s)	Input Price (1M)	Output Price (1M)
Qwen3 14B (Reasoning) Alibaba	16.2	77.4%	-	60.4%	0.99	63	$0.35	$4.20
Nanbeige4.1-3BNew Nanbeige	16.1	-	-	84.9%	-	-	$0.00	$0.00
Qwen: Qwen3 VL 30B A3B Instruct Alibaba	16.1	76.4%	131K	69.5%	0.93	127	$0.20	$0.80
DeepSeek: R1 Distill Llama 70B DeepSeek	16.0	79.5%	131K	40.2%	0.59	40	$0.70	$1.05
Gemini 1.5 Pro (Sep '24) Google	16.0	75.0%	-	58.9%	-	-	$0.00	$0.00
Hermes 4 - Llama-3.1 70B (Reasoning) Nous Research	16.0	81.1%	-	69.9%	0.58	81	$0.13	$0.40
Ministral 3 14B Mistral	16.0	69.3%	-	57.2%	0.33	110	$0.20	$0.20
Solar Pro 2 (Preview) (Non-reasoning) Upstage	16.0	72.5%	-	54.4%	-	-	$0.00	$0.00
Claude 3.5 Sonnet (Oct '24) Anthropic	15.9	77.2%	-	59.9%	-	-	$3.00	$15.00
DeepSeek R1 Distill Qwen 14B DeepSeek	15.8	74.0%	-	48.4%	-	-	$0.00	$0.00
Falcon-H1R-7BNew TII UAE	15.8	72.5%	-	66.1%	-	-	$0.00	$0.00
Ling-flash-2.0 InclusionAI	15.7	77.7%	-	65.7%	1.48	62	$0.14	$0.57
Qwen2.5 72B Instruct Alibaba	15.6	72.0%	33K	49.1%	1.08	54	$0.00	$0.00
Qwen3 Omni 30B A3B (Reasoning) Alibaba	15.6	79.2%	-	72.6%	0.95	103	$0.25	$0.97
Perplexity: Sonar Perplexity	15.5	68.9%	127K	47.1%	-	-	$0.00	$0.00
Step3 VL 10BNew StepFun	15.4	-	-	69.0%	-	-	$0.00	$0.00
Qwen: Qwen3 30B A3B Alibaba	15.3	77.7%	41K	61.6%	1.07	71	$0.20	$2.40
Gemma 4 E2B (Reasoning)New Google	15.2	-	-	43.3%	-	-	$0.00	$0.00
Mistral: Devstral Small 1.1 Mistral	15.2	62.2%	131K	41.4%	0.37	222	$0.10	$0.30
Perplexity: Sonar Pro Perplexity	15.2	75.5%	200K	57.8%	-	-	$0.00	$0.00

Showing 261-280 of 645 models