Best Agentic AI Models 2026

A definitive ranking of the top agentic LLMs and AI models optimized for your specific use cases.

Model	Intelligence	MMLU Pro	Context	GPQA	TTFT (s)	Speed (T/s)	Input Price (1M)	Output Price (1M)
Anthropic: Claude 3.5 Haiku anthropic	-	-	200K	-	-	-	$0.80	$4.00
Anthropic: Claude 3.5 Sonnet anthropic	-	-	200K	-	-	-	$6.00	$30.00
Google: Gemini 2.0 Flash google	-	-	1049K	-	-	-	$0.10	$0.40
Grok Voice Agent xAI	-	-	131K	-	-	-	$0.00	$0.00
Meta: Llama 3.2 3B Instruct meta-llama	-	-	131K	-	-	-	$0.02	$0.02
NVIDIA: Llama 3.3 Nemotron Super 49B V1.5 nvidia	-	-	131K	-	-	-	$0.10	$0.40
Qwen2.5 Coder 32B Instruct qwen	-	-	33K	-	-	-	$0.20	$0.20
Mistral: Mistral Small 3.1 24B mistralai	-	-	128K	-	-	-	$0.35	$0.56
Meituan: LongCat Flash Chat meituan	-	-	131K	-	-	-	$0.20	$0.80
NVIDIA: Nemotron 3 Nano 30B A3B nvidia	13.3	57.9%	262K	39.9%	0.23	120	$0.05	$0.20
Arcee AI: Trinity Mini arcee-ai	-	-	131K	-	-	-	$0.04	$0.15
Mistral: Mistral Small 3.2 24B mistralai	-	-	131K	-	-	-	$0.06	$0.18
Anthropic: Claude Opus 4 anthropic	-	-	200K	-	-	-	$15.00	$75.00
Anthropic: Claude Sonnet 4 anthropic	-	-	1000K	-	-	-	$3.00	$15.00
Anthropic: Claude 3.7 Sonnet anthropic	-	-	200K	-	-	-	$3.00	$15.00
Anthropic: Claude 3.7 Sonnet (thinking) anthropic	-	-	200K	-	-	-	$3.00	$15.00
Cohere: Command R7B (12-2024) cohere	-	-	128K	-	-	-	$0.04	$0.15
Nous: Hermes 3 405B Instruct nousresearch	-	-	131K	-	-	-	$1.00	$1.00
Anthropic: Claude Opus 4.5 anthropic	-	-	200K	-	1.36	78	$5.00	$25.00
Kwaipilot: KAT-Coder-Pro V1 kwaipilot	-	-	256K	-	1.98	56	$0.21	$0.83
Qwen: Qwen3 Next 80B A3B Thinking qwen	-	-	128K	-	-	-	$0.15	$1.20
Qwen: Qwen3 Next 80B A3B Instruct qwen	20.1	81.9%	262K	73.8%	1.06	131	$0.50	$2.00
Qwen: Qwen2.5-VL 7B Instruct qwen	-	-	33K	-	-	-	$0.20	$0.20
MiniMax: MiniMax M2.5 minimax	-	-	197K	-	-	-	$0.30	$1.10
Z.ai: GLM 4.7 Flash z-ai	-	-	203K	-	-	-	$0.06	$0.40
MiniMax: MiniMax M2.1 minimax	-	-	197K	-	-	-	$0.27	$0.95
Mistral: Mistral Small Creative mistralai	-	-	33K	-	-	-	$0.10	$0.30
Relace: Relace Search relace	-	-	256K	-	-	-	$1.00	$3.00
EssentialAI: Rnj 1 Instruct essentialai	-	-	33K	-	-	-	$0.15	$0.15
Perplexity: Sonar Pro Search perplexity	-	-	200K	-	-	-	$3.00	$15.00
Google: Gemini 3.1 Pro Preview google	-	-	1049K	-	35.19	92	$2.00	$12.00
MiniMax: MiniMax M2 minimax	36.0	82.0%	197K	77.7%	2.46	52	$0.30	$1.20
Qwen: Qwen3.5 397B A17B qwen	-	-	262K	-	1.44	74	$0.55	$3.50
MoonshotAI: Kimi K2.5New moonshotai	-	-	262K	-	1.28	41	$0.45	$2.20
OpenAI: gpt-oss-20b openai	24.5	74.8%	131K	68.8%	0.48	310	$0.07	$0.23
Google: Gemini 3 Flash Preview google	-	-	1049K	-	1.02	192	$0.50	$3.00
Qwen: Qwen3 30B A3B qwen	-	-	41K	-	1.00	69	$0.08	$0.28
Qwen: Qwen3 8B qwen	-	-	32K	-	0.94	64	$0.05	$0.40
Qwen: Qwen3 14B qwen	-	-	41K	-	0.98	53	$0.06	$0.24
Qwen: Qwen3 32B qwen	-	-	41K	-	0.99	88	$0.08	$0.24
Qwen: Qwen3 235B A22B qwen	-	-	131K	-	1.14	40	$0.46	$1.82
OpenAI: GPT-4.1 openai	-	-	1048K	-	0.52	86	$2.00	$8.00
Google: Gemini 3.1 Pro Preview Custom Tools google	-	-	1049K	-	-	-	$2.00	$12.00
Qwen: Qwen3 Max Thinking qwen	39.7	-	262K	86.1%	1.68	34	$1.20	$6.00
Qwen: Qwen3 Coder Next qwen	28.1	-	262K	73.7%	0.84	111	$0.20	$1.20
Xiaomi: MiMo-V2-Flash xiaomi	30.6	74.4%	262K	65.6%	1.22	146	$0.10	$0.30
Z.ai: GLM 5 z-ai	49.6	-	205K	82.0%	1.31	68	$1.00	$3.20
Mistral: Devstral 2 2512 mistralai	-	-	262K	-	-	-	$0.40	$2.00
Nex AGI: DeepSeek V3.1 Nex N1New nex-agi	-	-	131K	-	-	-	$0.27	$1.00
Writer: Palmyra X5 writer	-	-	1040K	-	-	-	$0.60	$6.00
Google: Gemini 3 Pro Preview google	-	-	1049K	-	-	-	$2.00	$12.00
Z.ai: GLM 4.7 z-ai	-	-	203K	-	-	-	$0.30	$1.40
Qwen: Qwen3 VL 32B Instruct qwen	17.2	79.1%	131K	67.1%	1.00	66	$0.70	$2.80
Qwen: Qwen3 VL 30B A3B Thinking qwen	-	-	131K	-	-	-	$0.00	$0.00
Amazon: Nova 2 Lite amazon	-	-	1000K	-	-	-	$0.30	$2.50
Qwen: Qwen3 VL 30B A3B Instruct qwen	16.0	76.4%	131K	69.5%	1.06	101	$0.20	$0.80
Qwen: Qwen3 Coder Plus qwen	-	-	1000K	-	-	-	$1.00	$5.00
Qwen: Qwen3 30B A3B Thinking 2507 qwen	-	-	33K	-	-	-	$0.05	$0.34
Mistral: Devstral Small 1.1 mistralai	15.2	62.2%	131K	41.4%	0.40	233	$0.10	$0.30
Inception: Mercury inception	-	-	128K	-	-	-	$0.25	$1.00
Qwen: Qwen3 VL 235B A22B Thinking qwen	-	-	131K	-	-	-	$0.00	$0.00
Qwen: Qwen3 VL 235B A22B Instruct qwen	20.6	82.3%	262K	71.2%	1.14	43	$0.70	$2.80
Qwen: Qwen3 Coder 30B A3B Instruct qwen	20.0	70.6%	160K	51.6%	1.46	21	$0.45	$2.25
Anthropic: Claude Haiku 4.5 anthropic	-	-	200K	-	-	-	$1.00	$5.00
Qwen: Qwen3 30B A3B Instruct 2507 qwen	-	-	262K	-	-	-	$0.09	$0.30
Qwen: Qwen3 Coder Flash qwen	-	-	1000K	-	-	-	$0.30	$1.50
Qwen: Qwen3 235B A22B Thinking 2507 qwen	-	-	131K	-	-	-	$0.00	$0.00
Z.ai: GLM 4.6 z-ai	-	-	203K	-	-	-	$0.35	$1.71
Qwen: Qwen3 Coder 480B A35B qwen	-	-	262K	-	-	-	$0.22	$1.00
Qwen: Qwen3 Coder 480B A35B (exacto) qwen	-	-	262K	-	-	-	$0.22	$1.80
Mistral: Devstral Medium mistralai	18.6	70.8%	131K	49.2%	0.43	106	$0.40	$2.00
OpenAI: o4 Mini High openai	-	-	200K	-	-	-	$1.10	$4.40
OpenAI: o4 Mini openai	33.0	83.2%	200K	78.4%	60.35	116	$1.10	$4.40
MoonshotAI: Kimi K2 ThinkingNew moonshotai	40.7	84.8%	131K	83.8%	0.65	72	$0.60	$2.50
OpenAI: GPT-5.2New openai	-	-	400K	-	0.60	73	$1.75	$14.00
DeepSeek: DeepSeek V3.2 ExpNew deepseek	-	-	164K	-	1.41	47	$0.27	$0.41
xAI: Grok Code Fast 1New x-ai	28.7	79.3%	256K	72.7%	7.16	268	$0.20	$1.50
OpenAI: GPT-5.3-CodexNew openai	-	-	400K	-	-	-	$1.75	$14.00
Anthropic: Claude Sonnet 4.6New anthropic	-	-	1000K	-	-	-	$3.00	$15.00
Anthropic: Claude Opus 4.6New anthropic	-	-	1000K	-	-	-	$5.00	$25.00
OpenAI: GPT-5.2-CodexNew openai	-	-	400K	-	-	-	$1.75	$14.00
OpenAI: GPT-5.1-Codex-MaxNew openai	-	-	400K	-	-	-	$1.25	$10.00
DeepSeek: DeepSeek V3.2 SpecialeNew deepseek	-	-	164K	-	-	-	$0.40	$1.20
DeepSeek: DeepSeek V3.2New deepseek	-	-	164K	-	1.29	46	$0.25	$0.40
OpenAI: GPT-5.1-CodexNew openai	-	-	400K	-	-	-	$1.25	$10.00
OpenAI: GPT-5.2 ProNew openai	-	-	400K	-	-	-	$21.00	$168.00
xAI: Grok 4.1 FastNew x-ai	-	-	2000K	-	0.79	118	$0.20	$0.50
Z.ai: GLM 4.5V z-ai	-	-	66K	-	-	-	$0.60	$1.80
OpenAI: GPT-5 CodexNew openai	44.5	86.5%	400K	83.7%	12.34	305	$1.25	$10.00
DeepSeek: DeepSeek V3.1 Terminus (exacto)New deepseek	-	-	164K	-	-	-	$0.21	$0.79
DeepSeek: DeepSeek V3.1 TerminusNew deepseek	-	-	164K	-	-	-	$0.21	$0.79
MoonshotAI: Kimi K2 0905New moonshotai	30.8	81.9%	131K	76.7%	0.50	63	$0.99	$2.50
MoonshotAI: Kimi K2 0905 (exacto)New moonshotai	-	-	262K	-	-	-	$0.60	$2.50
DeepSeek: DeepSeek V3.1New deepseek	-	-	33K	-	-	-	$0.15	$0.75
Z.ai: GLM 4.6 (exacto) z-ai	-	-	205K	-	-	-	$0.44	$1.76
Anthropic: Claude Sonnet 4.5 anthropic	-	-	1000K	-	-	-	$3.00	$15.00
MoonshotAI: Kimi K2 0711New moonshotai	26.2	82.4%	131K	76.6%	0.56	41	$0.60	$2.50
Qwen: Qwen2.5 Coder 7B Instruct qwen	-	-	33K	-	-	-	$0.03	$0.09
OpenAI: gpt-oss-120b openai	33.3	80.8%	131K	78.2%	0.48	314	$0.15	$0.60
Anthropic: Claude Opus 4.1 anthropic	-	-	200K	-	-	-	$15.00	$75.00
OpenAI: gpt-oss-120b (exacto) openai	-	-	131K	-	-	-	$0.04	$0.19
Cohere: Command R (08-2024) cohere	-	-	128K	-	-	-	$0.15	$0.60
Nous: Hermes 3 70B Instruct nousresearch	-	-	66K	-	-	-	$0.30	$0.30
Z.ai: GLM 4.5 z-ai	26.2	83.5%	131K	78.2%	0.90	40	$0.49	$1.90
Z.ai: GLM 4.5 Air z-ai	-	-	131K	-	-	-	$0.13	$0.85
Tongyi DeepResearch 30B A3B alibaba	-	-	131K	-	-	-	$0.09	$0.45
Nous: Hermes 4 70B nousresearch	-	-	131K	-	-	-	$0.13	$0.40
Nous: Hermes 4 405B nousresearch	-	-	131K	-	-	-	$1.00	$3.00
Z.ai: GLM 4 32B z-ai	-	-	128K	-	-	-	$0.10	$0.10
ByteDance: UI-TARS 7B bytedance	-	-	128K	-	-	-	$0.10	$0.20
MiniMax: MiniMax M1 minimax	-	-	1000K	-	-	-	$0.40	$2.20
Arcee AI: Spotlight arcee-ai	-	-	131K	-	-	-	$0.18	$0.18
Cohere: Command A cohere	13.4	71.2%	256K	52.7%	0.40	44	$2.50	$10.00