# Best LLMs for AI Agents in 2026

> AI models ranked for agentic work — tool use, browsing, computer use, and long-horizon task execution.

This page ranks model configurations by their agentic category score, using sourced terminal, browsing, computer-use, and tool-calling evidence. It helps choose a model for an agent whose model you can configure. It does not rank complete products such as Grok Bot, Hermes Agent, or OpenClaw.

Canonical page: https://benchlm.ai/best/agents

Last updated: October 13, 2026

## Rankings

| Rank | Model | Creator | Type | Context | Score |
|------|-------|---------|------|---------|-------|
| 1 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | Proprietary | 1M | 88.8 |
| 2 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | Proprietary | 1M | 79.3 |
| 3 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | Proprietary | null | 77.6 |
| 4 | [Claude Fable 5](/models/claude-fable) | Anthropic | Proprietary | 1M+ | 74.9 |
| 5 | [GPT-6.1 Sol](/models/gpt-6-1-sol) | OpenAI | Proprietary | 1.05M | 74.4 |
| 6 | [Gemini 4 Argon](/models/gemini-4-argon) | Google | Proprietary | null | 74.1 |
| 7 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | Proprietary | 1.05M | 70.7 |
| 8 | [Kimi K3](/models/kimi-k3) | Moonshot AI | Pending | 1.05M | 69 |
| 9 | [Claude Sonnet 5.5](/models/claude-sonnet-5-5) | Anthropic | Proprietary | 1M | 68.3 |
| 10 | [Grok 4.6](/models/grok-4-6) | xAI | Proprietary | 500K | 68.3 |
| 11 | [GLM-5.3](/models/glm-5-3) | Z.AI | Open Weight | 1M | 67.6 |
| 12 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | OpenAI | Proprietary | 1.05M | 67.4 |
| 13 | [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) | Xiaomi | Open Weight | 1M | 66 |
| 14 | [Qwen3.8 Max](/models/qwen3-8-max) | Alibaba | Open Weight | 1M | 65.9 |
| 15 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Google | Proprietary | 1M | 65.3 |
| 16 | [Claude Sonnet 5](/models/claude-sonnet-5) | Anthropic | Proprietary | 1M | 64.9 |
| 17 | [Step 5 Preview](/models/step-5-preview) | StepFun | Pending | 1M | 63.7 |
| 18 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | Proprietary | 1M | 62.3 |
| 19 | [Qwen3.8-27B](/models/qwen3-8-27b) | Alibaba | Open Weight | 262K | 61.7 |
| 20 | [Claude Haiku 5.5](/models/claude-haiku-5-5) | Anthropic | Proprietary | 1M | 61.7 |
| 21 | [GPT-5.5](/models/gpt-5-5) | OpenAI | Proprietary | 1M | 61.5 |
| 22 | [DeepSeek V4.1 Flash](/models/deepseek-v4-1-flash) | DeepSeek | Open Weight | 1M | 60.5 |
| 23 | [GPT-6 Sol](/models/gpt-6-sol) | OpenAI | Proprietary | 1.05M | 60.2 |
| 24 | [Claude Opus 4.7 (Adaptive)](/models/claude-opus-4-7-adaptive) | Anthropic | Proprietary | 1M | 59.8 |
| 25 | [Muse Spark 1.2](/models/muse-spark-1-2) | Meta | Proprietary | 1M | 59.4 |
| 26 | [Qwen3.8-Flash-Next](/models/qwen3-8-flash-next) | Alibaba | Open Weight | 262K | 59.1 |
| 27 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | OpenAI | Proprietary | 1.05M | 58.9 |
| 28 | [dots3-note Preview](/models/dots3-note-preview) | Dots Studio | Open Weight | 512K | 58.4 |
| 29 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | Google | Proprietary | 1M | 57.9 |
| 30 | [Ornith-1.5-397B](/models/ornith-1-5-397b) | Ornith AI | Open Weight | 262K | 57.6 |
| 31 | [Grok 4.5](/models/grok-4-5) | xAI | Proprietary | 500K | 57.5 |
| 32 | [Muse Spark 1.1](/models/muse-spark-1-1) | Meta | Proprietary | 1M | 57 |
| 33 | [GLM-5.2](/models/glm-5-2) | Z.AI | Open Weight | 1M | 56.5 |
| 34 | [GLM-5.3-Flash](/models/glm-5-3-flash) | Z.AI | Open Weight | 1M | 55.3 |
| 35 | [GPT-5.3 Codex](/models/gpt-5-3-codex) | OpenAI | Proprietary | 400K | 55.3 |
| 36 | [GPT-6 Luna](/models/gpt-6-luna) | OpenAI | Proprietary | 1.05M | 55 |
| 37 | [Hy4 preview](/models/hy4-preview) | Tencent | Open Weight | 1M | 54.5 |
| 38 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | OpenAI | Proprietary | 1.05M | 54.5 |
| 39 | [MiMo-V2.6-Flash](/models/mimo-v2-6-flash) | Xiaomi | Open Weight | 1M | 54.2 |
| 40 | [Claude Opus 4.7](/models/claude-opus-4-7) | Anthropic | Proprietary | 1M | 54 |
| 41 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | Open Weight | 1M | 53.2 |
| 42 | [Mistral Large 4](/models/mistral-large-4) | Mistral | Pending | 1M | 52.7 |
| 43 | [GPT-5.4](/models/gpt-5-4) | OpenAI | Proprietary | 1.05M | 51.3 |
| 44 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Google | Proprietary | 1M | 50.8 |
| 45 | [Muse Spark](/models/muse-spark) | Meta | Proprietary | 262K | 50.4 |
| 46 | [Agents-A1](/models/agents-a1) | InternScience | Open Weight | 262K | 47.6 |
| 47 | [Agents-A1-F16-GGUF](/models/agents-a1-f16-gguf) | InternScience | Open Weight | 262K | 47.6 |
| 48 | [Agents-A1-FP8](/models/agents-a1-fp8) | InternScience | Open Weight | 262K | 47.6 |
| 49 | [Agents-A1-Q4_K_M-GGUF](/models/agents-a1-q4-k-m-gguf) | InternScience | Open Weight | 262K | 47.6 |
| 50 | [Agents-A1-Q8_0-GGUF](/models/agents-a1-q8-0-gguf) | InternScience | Open Weight | 262K | 47.6 |
| 51 | [Claude Opus 4.6 (Adaptive)](/models/claude-opus-4-6-thinking) | Anthropic | Proprietary | 1M | 47.1 |
| 52 | [Apodex 1.1](/models/apodex-1-1) | Apodex | Proprietary | N/A | 45.8 |
| 53 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | Google | Proprietary | 1M | 45 |
| 54 | [Claude Opus 4.6](/models/claude-opus-4-6) | Anthropic | Proprietary | 1M | 44.6 |
| 55 | [Qwen3.5 397B](/models/qwen3-5-397b) | Alibaba | Open Weight | 128K | 44.3 |
| 56 | [GPT-5.2](/models/gpt-5-2) | OpenAI | Proprietary | 400K | 44 |
| 57 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | Open Weight | 256K | 43.8 |
| 58 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | Anthropic | Proprietary | 200K | 43.6 |
| 59 | [GLM-5.1](/models/glm-5-1) | Z.AI | Open Weight | 203K | 42.5 |
| 60 | [GLM-5](/models/glm-5) | Z.AI | Open Weight | 200K | 40.9 |
| 61 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | Open Weight | 256K | 40.2 |
| 62 | [MiniMax M3](/models/minimax-m3) | MiniMax | Open Weight | 1M | 40.2 |
| 63 | [GPT-5.2-Codex](/models/gpt-5-2-codex) | OpenAI | Proprietary | 400K | 40.1 |
| 64 | [Qwen3.7 Max](/models/qwen3-7-max) | Alibaba | Proprietary | 1M | 39.6 |
| 65 | [Quasar 438B](/models/quasar-438b) | Multiverse Computing | Proprietary | 1M | 39.1 |
| 66 | [Kimi K2.5 (Reasoning)](/models/kimi-k2-5-reasoning) | Moonshot AI | Proprietary | 128K | 38.8 |
| 67 | [Gemini 3.1 Pro](/models/gemini-3-1-pro) | Google | Proprietary | 1M | 38.4 |
| 68 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | Moonshot AI | Open Weight | 256K | 37.6 |
| 69 | [MiMo-V2.5-Pro](/models/mimo-v2-5-pro) | Xiaomi | Proprietary | 1M | 36.8 |
| 70 | [Qwen3.5-27B](/models/qwen3-5-27b) | Alibaba | Open Weight | 262K | 36.7 |
| 71 | [Hy3](/models/hy3) | Tencent | Open Weight | 256K | 36.2 |
| 72 | [Inkling](/models/inkling) | Thinking Machines Lab | Open Weight | 1M | 36.2 |
| 73 | [Step 3.7 Flash](/models/step-3-7-flash) | StepFun | Open Weight | 256K | 36.1 |
| 74 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | Open Weight | 1M | 35.8 |
| 75 | [Qwen3.7 Plus](/models/qwen3-7-plus) | Alibaba | Proprietary | 1M | 35.4 |
| 76 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | Proprietary | 400K | 34.7 |
| 77 | [Gemini 3.5 Flash-Lite](/models/gemini-3-5-flash-lite) | Google | Proprietary | 1M | 34.3 |
| 78 | [Claude Sonnet 4.5](/models/claude-sonnet-4-5) | Anthropic | Proprietary | 200K | 34.1 |
| 79 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | Proprietary | 1M | 33.5 |
| 80 | [Qwen3.5-35B-A3B](/models/qwen3-5-35b-a3b) | Alibaba | Open Weight | 262K | 33.1 |
| 81 | [Ling 3.0 Flash](/models/ling-3-0-flash) | InclusionAI | Open Weight | 262K | 32.5 |
| 82 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | Proprietary | 400K | 32.2 |
| 83 | [MiMo-V2.5](/models/mimo-v2-5) | Xiaomi | Proprietary | 1M | 32 |
| 84 | [GLM-4.7](/models/glm-4-7) | Z.AI | Open Weight | 200K | 31.8 |
| 85 | [Gemini 3 Flash](/models/gemini-3-flash) | Google | Proprietary | 1M | 31.7 |
| 86 | [MiniMax M2.7](/models/minimax-m2-7) | MiniMax | Open Weight | 200K | 31.4 |
| 87 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | Proprietary | 200K | 31.3 |
| 88 | [Grok Build 0.1](/models/grok-build-0-1) | xAI | Proprietary | 256K | 29.7 |
| 89 | [Qwen3.6-35B-A3B](/models/qwen3-6-35b-a3b) | Alibaba | Open Weight | 262K | 29.2 |
| 90 | [Qwen3.6-27B](/models/qwen3-6-27b) | Alibaba | Open Weight | 262K | 28.2 |
| 91 | [MiniMax M2.5](/models/minimax-m2-5) | MiniMax | Proprietary | 128K | 28.1 |
| 92 | [Ornith-1.5-35B-A3B](/models/ornith-1-5-35b-a3b) | Ornith AI | Open Weight | 262K | 27.9 |
| 93 | [Grok 4.3](/models/grok-4-3) | xAI | Proprietary | 1M | 27.5 |
| 94 | [Nemotron 3 Ultra](/models/nemotron-3-ultra) | NVIDIA | Open Weight | 1M | 27.2 |
| 95 | [GPT-5 mini](/models/gpt-5-mini) | OpenAI | Proprietary | 128K | 27.2 |
| 96 | [Gemini 3.1 Flash-Lite](/models/gemini-3-1-flash-lite) | Google | Proprietary | 1M | 26.7 |
| 97 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | Google | Proprietary | 1M | 25.2 |
| 98 | [Muse Glimmer 30B](/models/muse-glimmer-30b) | Meta | Open Weight | 131K | 24.5 |
| 99 | [Grok 4.20](/models/grok-4-20-beta) | xAI | Proprietary | 2M | 24.4 |
| 100 | [Ornith-1.5-9B](/models/ornith-1-5-9b) | Ornith AI | Open Weight | 262K | 23.6 |
| 101 | [Qwen3.5-122B-A10B](/models/qwen3-5-122b-a10b) | Alibaba | Open Weight | 262K | 23.3 |
| 102 | [Claude Haiku 4.5](/models/claude-haiku-4-5) | Anthropic | Proprietary | 200K | 20.8 |
| 103 | [Gemma 4 31B](/models/gemma-4-31b) | Google | Open Weight | 256K | 19.4 |
| 104 | [Mistral Medium 3.5 128B](/models/mistral-medium-3-5-128b) | Mistral | Open Weight | 256K | 19.2 |
| 105 | [GPT-OSS 120B](/models/gpt-oss-120b) | OpenAI | Open Weight | 128K | 19 |
| 106 | [Nemotron 3.5 Lightning 30B A3B NVFP4](/models/nemotron-3-5-lightning-30b-a3b-nvfp4) | NVIDIA | Open Weight | 1M | 16.7 |
| 107 | [Command A+](/models/command-a-plus) | Cohere | Open Weight | 128K | 16.4 |
| 108 | [LFM2.5-2.6B](/models/lfm2-5-2-6b) | LiquidAI | Open Weight | 128K | 16.3 |
| 109 | [Mercury 2](/models/mercury-2) | Inception | Proprietary | 128K | 15.3 |
| 110 | [Granite 4.2 8B](/models/granite-4-2-8b) | IBM | Open Weight | 128K | 15.3 |
| 111 | [GPT-OSS 20B](/models/gpt-oss-20b) | OpenAI | Open Weight | 128K | 14.6 |
| 112 | [Mistral Small 4](/models/mistral-small-4) | Mistral | Open Weight | 256K | 13.8 |
| 113 | [Trinity-Large-Thinking](/models/trinity-large-thinking) | Arcee AI | Open Weight | 512K | 13.2 |
| 114 | [Mercury 2.5](/models/mercury-2-5) | Inception | Proprietary | 260K | 11.8 |
| 115 | [Mistral Large 3](/models/mistral-large-3) | Mistral | Proprietary | 128K | 11 |
| 116 | [Llama 4 Maverick](/models/llama-4-maverick) | Meta | Open Weight | 1M | 10.8 |
| 117 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | Open Weight | 128K | 10 |
| 118 | [Llama 4 Scout](/models/llama-4-scout) | Meta | Open Weight | 10M | 7.7 |
| 119 | [Celeris-1](/models/celeris-1) | Celeris | Proprietary | 128K | 7.1 |
| 120 | [Ministral 3 14B](/models/ministral-3-14b) | Mistral | Open Weight | 128K | 6.9 |
| 121 | [Ministral 3 8B](/models/ministral-3-8b) | Mistral | Open Weight | 128K | 6.2 |
| 122 | [Gemma 3 27B](/models/gemma-3-27b) | Google | Open Weight | 32K | 5.9 |
| 123 | [Ministral 3 3B](/models/ministral-3-3b) | Mistral | Open Weight | 128K | 5.5 |
| — | [Mistral Large 4](/models/mistral-large-4) | Mistral | Pending | 1M | — |

## Key Takeaways

- Top model: [Claude Opus 5.5](/models/claude-opus-5-5) with a score of 88.8
- Best open-weight option: [GLM-5.3](/models/glm-5-3) at #11
- Models included: 124 (123 ranked, 1 listed without a rank)

## Compare the Leaders

- [Claude Opus 5.5 vs Claude Fable 5.1](/compare/claude-fable-5-1-vs-claude-opus-5-5)