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)
Claude 4 Opus (Reasoning) Anthropic	39.0	87.3%	-	79.6%	7.56	44	$15.00	$75.00
GPT-5 mini (medium) OpenAI	38.9	82.8%	-	80.3%	24.62	76	$0.25	$2.00
Claude 4 Sonnet (Reasoning) Anthropic	38.7	84.2%	-	77.7%	8.82	51	$3.00	$15.00
OpenAI: GPT-5.1-Codex-Mini OpenAI	38.6	82.0%	400K	81.3%	5.07	195	$0.25	$2.00
Qwen3.5 Omni PlusNew Alibaba	38.6	-	-	82.6%	1.25	48	$0.40	$4.80
xAI: Grok 4.1 Fast xAI	38.6	85.4%	2000K	85.3%	8.46	164	$0.20	$0.50
Step 3.5 Flash 2603New StepFun	38.5	-	-	82.6%	0.81	180	$0.00	$0.00
OpenAI: o3 OpenAI	38.4	85.3%	200K	82.7%	6.95	85	$2.00	$8.00
OpenAI: GPT-5.4 NanoNew OpenAI	38.1	-	400K	76.1%	2.57	156	$0.20	$1.25
StepFun: Step 3.5 FlashNew StepFun	37.8	-	262K	83.1%	0.84	167	$0.10	$0.30
OpenAI: GPT-5.4 MiniNew OpenAI	37.7	-	400K	82.3%	5.53	155	$0.75	$4.50
Kimi K2.5 (Non-reasoning)New Kimi	37.3	-	-	78.9%	1.42	31	$0.60	$3.00
Qwen3.5 27B (Non-reasoning)New Alibaba	37.2	-	-	84.2%	1.41	85	$0.30	$2.40
Claude 4.5 Haiku (Reasoning) Anthropic	37.1	76.0%	-	67.2%	10.91	133	$1.00	$5.00
Claude 4.5 Sonnet (Non-reasoning) Anthropic	37.1	86.0%	-	72.7%	0.89	48	$3.00	$15.00
Qwen: Qwen3.5-35B-A3BNew Alibaba	37.1	-	262K	84.5%	1.08	143	$0.25	$2.00
MiniMax: MiniMax M2 MiniMax	36.1	82.0%	197K	77.7%	1.64	62	$0.30	$1.20
Claude 4.1 Opus (Non-reasoning) Anthropic	36.0	-	-	-	1.33	35	$15.00	$75.00
KAT-Coder-Pro V1 KwaiKAT	36.0	81.3%	-	76.4%	1.17	113	$0.30	$1.20
NVIDIA: Nemotron 3 SuperNew NVIDIA	36.0	-	262K	80.0%	1.04	161	$0.09	$0.45

Showing 61-80 of 644 models