AI Agents

Models optimised for autonomous agent workflows.

Updated April 15, 2026
# Model Score AI Index Context Input / 1M Output / 1M Caps
1Google: Gemini 3.1 Pro Previewgoogle Top Pick95.357.21M$2.00$12.00TV
2Anthropic: Claude Opus 4.6anthropic Best for Coding In-House Pick94.2531M$5.00$25.00TV
3OpenAI: GPT-5.3-Codexopenai Top Pick9153.6400K$1.75$14.00TV
4Anthropic: Claude Sonnet 4.6anthropic In-House Pick86.651.71M$3.00$15.00TV
5Google: Gemini 3 Flash Previewgoogle84.146.41M$0.5000$3.00TV
6OpenAI: GPT-5.1openai83.447.7400K$1.25$10.00TV
7Anthropic: Claude Opus 4.5anthropic83.349.7200K$5.00$25.00TV
8xAI: Grok 4.20x-ai New82.949.32M$2.00$6.00TV
9xAI: Grok 4.20 Multi-Agent Betax-ai82.748.52M$2.00$6.00TV
10OpenAI: GPT-5.2-Codexopenai81.649400K$1.75$14.00TV
11Xiaomi: MiMo-V2-Proxiaomi New81.549.21M$1.00$3.00T
12Google: Nano Banana Pro (Gemini 3 Pro Image Preview)google80.448.466K$2.00$12.00V
13MiniMax: MiniMax M2.7minimax New Top Pick79.649.6197K$0.3000$1.20T
14OpenAI: GPT-5 Codexopenai79.144.6400K$1.25$10.00TV
15Z.ai: GLM 5z-ai77.740.680K$0.7200$2.30T
16Google: Gemini 3 Pro Previewgoogle77.641.31M$2.00$12.00TV
17OpenAI: GPT-5openai77.623.9400K$1.25$10.00TV
18Kwaipilot: KAT-Coder-Pro V2kwaipilot New Best for Agents77.243.8256K$0.3000$1.20T
19Google: Gemini 2.5 Pro Preview 06-05google77.130.31M$1.25$10.00TV
20Anthropic: Claude Opus 4.1anthropic76.942200K$15.00$75.00TV
21Google: Gemma 4 31Bgoogle New76.839.2262K$0.1300$0.3800TV
22OpenAI: GPT-5.1-Codexopenai76.543.1400K$1.25$10.00TV
23Anthropic: Claude Sonnet 4.5anthropic76.3431M$3.00$15.00TV
24OpenAI: GPT-5 Miniopenai76.341.2400K$0.2500$2.00TV
25Anthropic: Claude Opus 4anthropic74.939200K$15.00$75.00TV
26OpenAI: o3openai74.838.4200K$2.00$8.00TV
27Xiaomi: MiMo-V2-Omnixiaomi New74.743.4262K$0.4000$2.00TV
28Z.ai: GLM 5 Turboz-ai74.646.8203K$1.20$4.00T
29OpenAI: GPT-5.1-Codex-Miniopenai74.438.6400K$0.2500$2.00TV
30Z.ai: GLM 4.7z-ai74.134.2203K$0.3900$1.75T
31OpenAI: GPT-5.4 Nanoopenai New74.124.4400K$0.2000$1.25TV
32MiniMax: MiniMax M2.5minimax73.541.9197K$0.1180$0.9900T
33xAI: Grok 4x-ai73.341.5256K$3.00$15.00TV
34Qwen: Qwen3.5-122B-A10Bqwen72.935.9262K$0.2600$2.08TV
35Qwen: Qwen3.5 397B A17Bqwen72.740.1262K$0.3900$2.34TV
36OpenAI: GPT-5.4 Miniopenai New Best for Coding72.548.9400K$0.7500$4.50TV
37Qwen: Qwen3.5-27Bqwen72.437.2262K$0.1950$1.56TV
38Z.ai: GLM 5V Turboz-ai New7242.9203K$1.20$4.00TV
39OpenAI: o3 Deep Researchopenai71.838.3200K$10.00$40.00TV
40Google: Gemini 2.5 Progoogle71.534.61M$1.25$10.00TV
41MiniMax: MiniMax M2.1minimax70.839.4197K$0.2900$0.9500T
42MoonshotAI: Kimi K2 Thinkingmoonshotai70.640.9262K$0.6000$2.50T
43Anthropic: Claude Sonnet 4anthropic70.538.71M$3.00$15.00TV
44Anthropic: Claude 3.7 Sonnet (thinking)anthropic69.634.7200K$3.00$15.00TV
45StepFun: Step 3.5 Flashstepfun Best Value68.737.8262K$0.1000$0.3000T
46MiniMax: MiniMax M2minimax68.536.1197K$0.2550$1.00T
47NVIDIA: Nemotron 3 Supernvidia66.736262K$0.1000$0.5000T
48DeepSeek: DeepSeek V3.2deepseek66.341.7164K$0.2600$0.3800T
49Anthropic: Claude 3.7 Sonnetanthropic65.930.8200K$3.00$15.00TV
50Kwaipilot: KAT-Coder-Pro V1kwaipilot64.536256K$0.2070$0.8280T
#1Top Pick95.3
Google: Gemini 3.1 Pro Previewgoogle
AI 57.21M ctx$2.00/M in
Tool UseVision
#2Best for CodingIn-House Pick94.2
Anthropic: Claude Opus 4.6anthropic
AI 531M ctx$5.00/M in
Tool UseVision
#3Top Pick91
OpenAI: GPT-5.3-Codexopenai
AI 53.6400K ctx$1.75/M in
Tool UseVision
#4In-House Pick86.6
Anthropic: Claude Sonnet 4.6anthropic
AI 51.71M ctx$3.00/M in
Tool UseVision
#584.1
Google: Gemini 3 Flash Previewgoogle
AI 46.41M ctx$0.5000/M in
Tool UseVision
#683.4
OpenAI: GPT-5.1openai
AI 47.7400K ctx$1.25/M in
Tool UseVision
#783.3
Anthropic: Claude Opus 4.5anthropic
AI 49.7200K ctx$5.00/M in
Tool UseVision
#8New82.9
xAI: Grok 4.20x-ai
AI 49.32M ctx$2.00/M in
Tool UseVision
#982.7
xAI: Grok 4.20 Multi-Agent Betax-ai
AI 48.52M ctx$2.00/M in
Tool UseVision
#1081.6
OpenAI: GPT-5.2-Codexopenai
AI 49400K ctx$1.75/M in
Tool UseVision
#11New81.5
Xiaomi: MiMo-V2-Proxiaomi
AI 49.21M ctx$1.00/M in
Tool Use
#1280.4
Google: Nano Banana Pro (Gemini 3 Pro Image Preview)google
AI 48.466K ctx$2.00/M in
Vision
#13NewTop Pick79.6
MiniMax: MiniMax M2.7minimax
AI 49.6197K ctx$0.3000/M in
Tool Use
#1479.1
OpenAI: GPT-5 Codexopenai
AI 44.6400K ctx$1.25/M in
Tool UseVision
#1577.7
Z.ai: GLM 5z-ai
AI 40.680K ctx$0.7200/M in
Tool Use
#1677.6
Google: Gemini 3 Pro Previewgoogle
AI 41.31M ctx$2.00/M in
Tool UseVision
#1777.6
OpenAI: GPT-5openai
AI 23.9400K ctx$1.25/M in
Tool UseVision
#18NewBest for Agents77.2
Kwaipilot: KAT-Coder-Pro V2kwaipilot
AI 43.8256K ctx$0.3000/M in
Tool Use
#1977.1
Google: Gemini 2.5 Pro Preview 06-05google
AI 30.31M ctx$1.25/M in
Tool UseVision
#2076.9
Anthropic: Claude Opus 4.1anthropic
AI 42200K ctx$15.00/M in
Tool UseVision
#21New76.8
Google: Gemma 4 31Bgoogle
AI 39.2262K ctx$0.1300/M in
Tool UseVision
#2276.5
OpenAI: GPT-5.1-Codexopenai
AI 43.1400K ctx$1.25/M in
Tool UseVision
#2376.3
Anthropic: Claude Sonnet 4.5anthropic
AI 431M ctx$3.00/M in
Tool UseVision
#2476.3
OpenAI: GPT-5 Miniopenai
AI 41.2400K ctx$0.2500/M in
Tool UseVision
#2574.9
Anthropic: Claude Opus 4anthropic
AI 39200K ctx$15.00/M in
Tool UseVision
#2674.8
OpenAI: o3openai
AI 38.4200K ctx$2.00/M in
Tool UseVision
#27New74.7
Xiaomi: MiMo-V2-Omnixiaomi
AI 43.4262K ctx$0.4000/M in
Tool UseVision
#2874.6
Z.ai: GLM 5 Turboz-ai
AI 46.8203K ctx$1.20/M in
Tool Use
#2974.4
OpenAI: GPT-5.1-Codex-Miniopenai
AI 38.6400K ctx$0.2500/M in
Tool UseVision
#3074.1
Z.ai: GLM 4.7z-ai
AI 34.2203K ctx$0.3900/M in
Tool Use
#31New74.1
OpenAI: GPT-5.4 Nanoopenai
AI 24.4400K ctx$0.2000/M in
Tool UseVision
#3273.5
MiniMax: MiniMax M2.5minimax
AI 41.9197K ctx$0.1180/M in
Tool Use
#3373.3
xAI: Grok 4x-ai
AI 41.5256K ctx$3.00/M in
Tool UseVision
#3472.9
Qwen: Qwen3.5-122B-A10Bqwen
AI 35.9262K ctx$0.2600/M in
Tool UseVision
#3572.7
Qwen: Qwen3.5 397B A17Bqwen
AI 40.1262K ctx$0.3900/M in
Tool UseVision
#36NewBest for Coding72.5
OpenAI: GPT-5.4 Miniopenai
AI 48.9400K ctx$0.7500/M in
Tool UseVision
#3772.4
Qwen: Qwen3.5-27Bqwen
AI 37.2262K ctx$0.1950/M in
Tool UseVision
#38New72
Z.ai: GLM 5V Turboz-ai
AI 42.9203K ctx$1.20/M in
Tool UseVision
#3971.8
OpenAI: o3 Deep Researchopenai
AI 38.3200K ctx$10.00/M in
Tool UseVision
#4071.5
Google: Gemini 2.5 Progoogle
AI 34.61M ctx$1.25/M in
Tool UseVision
#4170.8
MiniMax: MiniMax M2.1minimax
AI 39.4197K ctx$0.2900/M in
Tool Use
#4270.6
MoonshotAI: Kimi K2 Thinkingmoonshotai
AI 40.9262K ctx$0.6000/M in
Tool Use
#4370.5
Anthropic: Claude Sonnet 4anthropic
AI 38.71M ctx$3.00/M in
Tool UseVision
#4469.6
Anthropic: Claude 3.7 Sonnet (thinking)anthropic
AI 34.7200K ctx$3.00/M in
Tool UseVision
#45Best Value68.7
StepFun: Step 3.5 Flashstepfun
AI 37.8262K ctx$0.1000/M in
Tool Use
#4668.5
MiniMax: MiniMax M2minimax
AI 36.1197K ctx$0.2550/M in
Tool Use
#4766.7
NVIDIA: Nemotron 3 Supernvidia
AI 36262K ctx$0.1000/M in
Tool Use
#4866.3
DeepSeek: DeepSeek V3.2deepseek
AI 41.7164K ctx$0.2600/M in
Tool Use
#4965.9
Anthropic: Claude 3.7 Sonnetanthropic
AI 30.8200K ctx$3.00/M in
Tool UseVision
#5064.5
Kwaipilot: KAT-Coder-Pro V1kwaipilot
AI 36256K ctx$0.2070/M in
Tool Use