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)
GLM-5 (Reasoning)New Z AI	49.8	-	-	82.0%	0.86	67	$1.00	$3.20
Claude Opus 4.5 (Reasoning) Anthropic	49.7	89.5%	-	86.6%	10.52	67	$5.00	$25.00
MiniMax: MiniMax M2.7New MiniMax	49.6	-	197K	87.4%	1.63	50	$0.30	$1.20
xAI: Grok 4.20New xAI	49.3	-	2000K	91.1%	21.53	96	$2.00	$6.00
Xiaomi: MiMo-V2-ProNew Xiaomi	49.2	-	1049K	87.0%	-	-	$0.00	$0.00
OpenAI: GPT-5.2-Codex OpenAI	49.0	-	400K	89.9%	8.62	105	$1.75	$14.00
GPT-5.4 mini (xhigh)New OpenAI	48.9	-	-	87.5%	5.09	176	$0.75	$4.50
Grok 4.20 0309 (Reasoning)New xAI	48.5	-	-	88.5%	32.15	97	$2.00	$6.00
Gemini 3 Pro Preview (high) Google	48.4	89.8%	-	90.8%	33.27	131	$2.00	$12.00
GPT-5.1 (high) OpenAI	47.7	87.0%	-	87.3%	26.61	146	$1.25	$10.00
Kimi K2.5 (Reasoning)New Kimi	46.8	-	-	87.9%	1.44	33	$0.60	$3.00
Z.ai: GLM 5 TurboNew Z AI	46.8	-	203K	84.7%	-	-	$0.00	$0.00
GPT-5.2 (medium) OpenAI	46.6	85.9%	-	86.4%	-	-	$1.75	$14.00
Anthropic: Claude Opus 4.6New Anthropic	46.5	-	1000K	84.0%	1.49	53	$5.00	$25.00
DeepSeek: DeepSeek V4 FlashNew DeepSeek	46.5	-	1049K	89.4%	0.83	83	$0.14	$0.28
Google: Gemini 3 Flash Preview Google	46.4	89.0%	1049K	89.8%	5.74	201	$0.50	$3.00
Qwen3.6 27B (Reasoning)New Alibaba	45.8	-	-	84.2%	1.73	68	$0.60	$3.60
Qwen: Qwen3.5 397B A17BNew Alibaba	45.0	-	262K	89.3%	1.54	51	$0.60	$3.60
DeepSeek V4 Flash (Reasoning, High Effort)New DeepSeek	44.9	-	-	86.7%	-	-	$0.14	$0.28
MiMo-V2-Omni-0327New Xiaomi	44.9	-	-	85.5%	-	-	$0.00	$0.00

Showing 21-40 of 667 models