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)
Nova Premier Amazon	19.0	73.3%	-	56.9%	1.14	53	$2.50	$12.50
DeepSeek R1 (Jan '25) DeepSeek	18.8	84.4%	-	70.8%	-	-	$1.35	$3.50
Gemma 4 E4B (Reasoning)New Google	18.8	-	-	57.6%	-	-	$0.00	$0.00
Magistral Medium 1 Mistral	18.8	75.3%	-	67.9%	-	-	$0.00	$0.00
Mistral: Mistral Medium 3 Mistral	18.8	76.0%	131K	57.8%	0.45	52	$0.40	$2.00
Solar Pro 2 (Preview) (Reasoning) Upstage	18.8	76.8%	-	57.8%	-	-	$0.00	$0.00
Anthropic: Claude 3.5 Haiku Anthropic	18.7	63.4%	200K	40.8%	-	-	$0.80	$4.00
K2-V2 (medium) MBZUAI Institute of Foundation Models	18.7	76.1%	-	59.8%	-	-	$0.00	$0.00
Llama Nemotron Super 49B v1.5 (Reasoning) NVIDIA	18.7	81.4%	-	74.8%	0.35	62	$0.10	$0.40
Mistral: Devstral Medium Mistral	18.7	70.8%	131K	49.2%	0.46	142	$0.40	$2.00
GPT-4o (March 2025, chatgpt-4o-latest) OpenAI	18.6	80.3%	-	65.5%	-	-	$0.00	$0.00
Hermes 4 - Llama-3.1 405B (Reasoning) Nous Research	18.6	82.9%	-	72.7%	0.78	35	$1.00	$3.00
Mistral: Mistral Small 4New Mistral	18.6	-	262K	57.1%	0.51	148	$0.15	$0.60
OpenAI: GPT-4o (2024-08-06) OpenAI	18.6	-	128K	52.1%	0.56	98	$2.50	$10.00
Tri-21B-ThinkNew Trillion Labs	18.6	-	-	60.1%	-	-	$0.00	$0.00
Gemini 2.0 Flash (Feb '25) Google	18.5	77.9%	-	62.3%	-	-	$0.15	$0.60
Llama 3.3 Nemotron Super 49B v1 (Reasoning) NVIDIA	18.5	78.5%	-	64.3%	-	-	$0.00	$0.00
Meta: Llama 4 Maverick Meta	18.4	80.9%	1049K	67.1%	0.61	115	$0.35	$0.85
Magistral Small 1.2 Mistral	18.2	76.8%	-	66.3%	0.35	179	$0.50	$1.50
Qwen3 4B 2507 (Reasoning) Alibaba	18.2	74.3%	-	66.7%	-	-	$0.00	$0.00

Showing 221-240 of 644 models