# Best AI Models in 2026 — Overall Rankings

> The top AI models ranked by overall benchmark performance across all categories.

BenchLM.ai now distinguishes provisional overall ranking from verified overall ranking. The provisional score is a normalized weighted average across 8 benchmark categories: agentic (22%), coding (20%), reasoning (17%), knowledge (12%), multimodal & grounded (12%), multilingual (7%), instruction following (5%), and math (5%), using non-generated benchmark coverage plus bounded external consensus calibration. The verified leaderboard is stricter and only counts sourced benchmark rows. Each score includes a confidence indicator (1-4 dots) based on how much sourced coverage supports it. Display-only benchmarks — including MMLU, OpenBookQA, HumanEval, FLTEval, BBH, LisanBench, and older AIME/HMMT variants — remain visible for context but do not affect ranking.

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

Last updated: October 17, 2026

## Rankings

Estimated ranges describe retrospective reconstruction error. Sparse source patterns lack validated coverage; no range establishes confidence in a rank or future score. See the [methodology limits](/methodology#estimate-validation-limits).

| Rank | Model | Creator | Type | Context | Score | Evidence | Conditional range |
|------|-------|---------|------|---------|-------|----------|--------------|
| 1 | [Claude Opus 5.5](/models/claude-opus-5-5) | Anthropic | Proprietary | 1M | 86.6 | Supported | 81.74–91.38 |
| 2 | [GPT-6.1 Sol](/models/gpt-6-1-sol) | OpenAI | Proprietary | 1.05M | 84.1 | Supported | 80.10–88.04 |
| 3 | [GPT-6 Astra](/models/gpt-6-astra) | OpenAI | Proprietary | 1.05M | 83.9 | Supported | 79.41–88.35 |
| 4 | [Claude Fable 5.1](/models/claude-fable-5-1) | Anthropic | Proprietary | 1M | 83.2 | Supported | 79.90–86.46 |
| 5 | [Claude Sonnet 5.5](/models/claude-sonnet-5-5) | Anthropic | Proprietary | 1M | 82.2 | Supported | 76.53–87.84 |
| 6 | [Claude Fable 5](/models/claude-fable) | Anthropic | Proprietary | 1M+ | 81.3 | Supported | 78.35–84.18 |
| 7 | [Gemini 4 Argon](/models/gemini-4-argon) | Google | Proprietary | 1M | 81 | Supported | 78.14–83.79 |
| 8 | [Claude Opus 5](/models/claude-opus-5) | Anthropic | Proprietary | null | 79.6 | Supported | 76.83–82.43 |
| 9 | [GPT-6 Sol](/models/gpt-6-sol) | OpenAI | Proprietary | 1.05M | 78.5 | Supported | 75.27–81.68 |
| 10 | [GPT-5.6 Sol](/models/gpt-5-6-sol) | OpenAI | Proprietary | 1.05M | 75.3 | Supported | 72.61–78.03 |
| 11 | [Muse Wonder Full](/models/muse-wonder-full) | Meta | Proprietary | 1M | 74.3 | Estimated | 59.92–88.62 |
| 12 | [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) | Xiaomi | Open Weight | 1M | 73.6 | Estimated | 63.85–83.29 |
| 13 | [GPT-5.5 Pro](/models/gpt-5-5-pro) | OpenAI | Proprietary | 1M | 73.4 | Estimated | 59.02–87.72 |
| 14 | [Muse Wonder](/models/muse-wonder) | Meta | Proprietary | 1M | 71.2 | Estimated | 56.81–85.51 |
| 15 | [GPT-5.6 Terra](/models/gpt-5-6-terra) | OpenAI | Proprietary | 1.05M | 70.2 | Supported | 66.51–73.91 |
| 16 | [Claude Haiku 5.5](/models/claude-haiku-5-5) | Anthropic | Proprietary | 1M | 70 | Supported | 65.93–74.16 |
| 17 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | Proprietary | 1M | 70 | Supported | 66.14–73.88 |
| 18 | [Kimi K3](/models/kimi-k3) | Moonshot AI | Pending | 1.05M | 69.9 | Supported | 65.75–73.99 |
| 19 | [Qwen3.8 Max](/models/qwen3-8-max) | Alibaba | Open Weight | 1M | 69.7 | Supported | 65.60–73.73 |
| 20 | [Step 5 Preview](/models/step-5-preview) | StepFun | Open Weight | 1M | 69.4 | Estimated | 59.71–79.16 |
| 21 | [GPT-5.5](/models/gpt-5-5) | OpenAI | Proprietary | 1M | 69.3 | Supported | 62.97–75.61 |
| 22 | [Gemini 3.8 Flash](/models/gemini-3-8-flash) | Google | Proprietary | 1M | 69.3 | Supported | 63.63–74.87 |
| 23 | [Claude Opus 4.7 (Adaptive)](/models/claude-opus-4-7-adaptive) | Anthropic | Proprietary | 1M | 69 | Estimated | 54.61–83.31 |
| 24 | [GPT-5.4 Pro](/models/gpt-5-4-pro) | OpenAI | Proprietary | 1.05M | 68.6 | Estimated | 54.25–82.95 |
| 25 | [Ling 3.1 Flash](/models/ling-3-1-flash) | InclusionAI | Open Weight | 262K | 68 | Estimated | 53.68–82.38 |
| 26 | [GLM-5.3](/models/glm-5-3) | Z.AI | Open Weight | 1M | 67.9 | Supported | 63.43–72.34 |
| 27 | [Grok 4.6](/models/grok-4-6) | xAI | Proprietary | 500K | 67 | Supported | 63.72–70.31 |
| 28 | [Gemini 3.7 Flash](/models/gemini-3-7-flash) | Google | Proprietary | 1M | 66.8 | Supported | 61.22–72.46 |
| 29 | [Claude Opus 4.7](/models/claude-opus-4-7) | Anthropic | Proprietary | 1M | 66.7 | Estimated | 57.02–76.46 |
| 30 | [GPT-5.4](/models/gpt-5-4) | OpenAI | Proprietary | 1.05M | 66.7 | Supported | 62.22–71.18 |
| 31 | [GPT-6 Luna](/models/gpt-6-luna) | OpenAI | Proprietary | 1.05M | 66.7 | Supported | 62.49–70.82 |
| 32 | [GPT-5.2 Pro](/models/gpt-5-2-pro) | OpenAI | Proprietary | 400K | 66.7 | Estimated | 56.93–76.38 |
| 33 | [Muse Spark 1.1](/models/muse-spark-1-1) | Meta | Proprietary | 1M | 66 | Supported | 57.51–74.53 |
| 34 | [Muse Spark 1.2](/models/muse-spark-1-2) | Meta | Proprietary | 1M | 65.8 | Supported | 59.70–71.99 |
| 35 | [DeepSeek V4.1 Flash 1017](/models/deepseek-v4-1-flash) | DeepSeek | Open Weight | 1M | 65.6 | Supported | 60.59–70.63 |
| 36 | [Claude Sonnet 5](/models/claude-sonnet-5) | Anthropic | Proprietary | 1M | 64.8 | Supported | 59.86–69.82 |
| 37 | [Gemini 3.1 Pro](/models/gemini-3-1-pro) | Google | Proprietary | 1M | 64.7 | Estimated | 54.98–74.43 |
| 38 | [MiMo-V2.6-Flash](/models/mimo-v2-6-flash) | Xiaomi | Open Weight | 1M | 64.5 | Estimated | 54.79–74.23 |
| 39 | [GPT-5.6 Luna](/models/gpt-5-6-luna) | OpenAI | Proprietary | 1.05M | 64.4 | Supported | 59.28–69.42 |
| 40 | [dots3-note Preview](/models/dots3-note-preview) | Dots Studio | Open Weight | 512K | 63.8 | Estimated | 49.43–78.13 |
| 41 | [Grok 4.5](/models/grok-4-5) | xAI | Proprietary | 500K | 63.3 | Supported | 57.92–68.63 |
| 42 | [DeepSeek V4 Pro 0813](/models/deepseek-v4-pro-0813) | DeepSeek | Open Weight | 1M | 63.2 | Estimated | 48.84–77.54 |
| 43 | [Qwen3.8-Flash-Next](/models/qwen3-8-flash-next) | Alibaba | Open Weight | 262K | 62.8 | Estimated | 48.45–77.15 |
| 44 | [Gemini 3.6 Flash](/models/gemini-3-6-flash) | Google | Proprietary | 1M | 62.8 | Supported | 55.07–70.49 |
| 45 | [Muse Spark Open](/models/muse-spark-open) | Meta | Open Weight | 1M | 62.7 | Estimated | 48.31–77.01 |
| 46 | [Qwen3.7 Max](/models/qwen3-7-max) | Alibaba | Proprietary | 1M | 62.6 | Supported | 52.82–72.34 |
| 47 | [Claude Opus 4.6](/models/claude-opus-4-6) | Anthropic | Proprietary | 1M | 62.4 | Supported | 50.18–74.61 |
| 48 | [Ornith-1.5-397B](/models/ornith-1-5-397b) | Ornith AI | Open Weight | 262K | 62.3 | Estimated | 47.98–76.68 |
| 49 | [Gemini 3.5 Flash](/models/gemini-3-5-flash) | Google | Proprietary | 1M | 61.9 | Supported | 53.66–70.11 |
| 50 | [Mistral Large 4](/models/mistral-large-4) | Mistral | Pending | 1M | 61.5 | Supported | 54.44–68.59 |
| 51 | [Hy4 preview](/models/hy4-preview) | Tencent | Open Weight | 1M | 61.3 | Estimated | 46.92–75.62 |
| 52 | [Gemini 3 Pro](/models/gemini-3-pro) | Google | Proprietary | 2M | 61.2 | Supported | 50.33–72.03 |
| 53 | [GLM-5.2](/models/glm-5-2) | Z.AI | Open Weight | 1M | 60.5 | Supported | 52.56–68.45 |
| 54 | [GPT-5.3 Codex](/models/gpt-5-3-codex) | OpenAI | Proprietary | 400K | 60.5 | Supported | 53.11–67.85 |
| 55 | [Muse Spark](/models/muse-spark) | Meta | Proprietary | 262K | 60.5 | Supported | 50.61–70.33 |
| 56 | [GPT-5.2](/models/gpt-5-2) | OpenAI | Proprietary | 400K | 59.9 | Supported | 52.74–66.98 |
| 57 | [Qwen3.8-27B](/models/qwen3-8-27b) | Alibaba | Open Weight | 262K | 59.6 | Supported | 53.68–65.54 |
| 58 | [GLM-5.3-Flash](/models/glm-5-3-flash) | Z.AI | Open Weight | 1M | 58.5 | Estimated | 48.79–68.24 |
| 59 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | Open Weight | 256K | 58.4 | Estimated | 48.68–68.13 |
| 60 | [Grok 4.20](/models/grok-4-20-beta) | xAI | Proprietary | 2M | 57.4 | Estimated | 47.71–67.15 |
| 61 | [GPT-5.1](/models/gpt-5-1) | OpenAI | Proprietary | 200K | 57.2 | Supported | 46.44–67.98 |
| 62 | [GLM-5.1](/models/glm-5-1) | Z.AI | Open Weight | 203K | 56.3 | Supported | 45.58–67.03 |
| 63 | [Kimi K2.5 (Reasoning)](/models/kimi-k2-5-reasoning) | Moonshot AI | Proprietary | 128K | 56.1 | Estimated | 41.74–70.43 |
| 64 | [Claude Sonnet 4.6](/models/claude-sonnet-4-6) | Anthropic | Proprietary | 200K | 55.9 | Supported | 43.60–68.19 |
| 65 | [Qwen3.5 397B](/models/qwen3-5-397b) | Alibaba | Open Weight | 128K | 55.3 | Estimated | 40.99–69.69 |
| 66 | [GPT-5.2-Codex](/models/gpt-5-2-codex) | OpenAI | Proprietary | 400K | 55.3 | Estimated | 45.58–65.03 |
| 67 | [Qwen 3.6 Max (preview)](/models/qwen3-6-max-preview) | Alibaba | Proprietary | 256K | 55.1 | Estimated | 45.36–64.80 |
| 68 | [MiniMax M3](/models/minimax-m3) | MiniMax | Open Weight | 1M | 54.9 | Supported | 45.49–64.31 |
| 69 | [GLM-5](/models/glm-5) | Z.AI | Open Weight | 200K | 54.8 | Supported | 44.15–65.52 |
| 70 | [Gemini 3 Flash](/models/gemini-3-flash) | Google | Proprietary | 1M | 54.3 | Supported | 39.10–69.46 |
| 71 | [Agents-A1](/models/agents-a1) | InternScience | Open Weight | 262K | 54.2 | Estimated | 39.85–68.55 |
| 72 | [Agents-A1-F16-GGUF](/models/agents-a1-f16-gguf) | InternScience | Open Weight | 262K | 54.2 | Estimated | 39.85–68.55 |
| 73 | [Agents-A1-FP8](/models/agents-a1-fp8) | InternScience | Open Weight | 262K | 54.2 | Estimated | 39.85–68.55 |
| 74 | [Agents-A1-Q4_K_M-GGUF](/models/agents-a1-q4-k-m-gguf) | InternScience | Open Weight | 262K | 54.2 | Estimated | 39.85–68.55 |
| 75 | [Agents-A1-Q8_0-GGUF](/models/agents-a1-q8-0-gguf) | InternScience | Open Weight | 262K | 54.2 | Estimated | 39.85–68.55 |
| 76 | [Claude Opus 4.5 Thinking](/models/claude-opus-4-5-thinking) | Anthropic | Proprietary | 200K | 54.1 | Estimated | 39.75–68.45 |
| 77 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | Proprietary | 200K | 54.1 | Supported | 41.41–66.68 |
| 78 | [Ternary Bonsai 2 27B](/models/ternary-bonsai-2-27b) | Prism ML | Open Weight | 262K | 53.9 | Estimated | 39.52–68.22 |
| 79 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | Proprietary | 1M | 53.8 | Supported | 44.17–63.40 |
| 80 | [Qwen3.7 Plus](/models/qwen3-7-plus) | Alibaba | Proprietary | 1M | 53.8 | Supported | 42.68–64.84 |
| 81 | [Inkling-Small](/models/inkling-small) | Thinking Machines Lab | Open Weight | 1M | 53.5 | Supported | 45.05–61.94 |
| 82 | [GPT-5.4 mini](/models/gpt-5-4-mini) | OpenAI | Proprietary | 400K | 53.4 | Supported | 42.63–64.19 |
| 83 | [GLM-5-Turbo](/models/glm-5-turbo) | Z.AI | Proprietary | 200K | 53.2 | Estimated | 43.46–62.91 |
| 84 | [MAI-Thinking-1](/models/mai-thinking-1) | Microsoft | Proprietary | 256K | 53.2 | Estimated | 38.82–67.52 |
| 85 | [Grok 4.3](/models/grok-4-3) | xAI | Proprietary | 1M | 53.1 | Supported | 42.96–63.18 |
| 86 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | Open Weight | 256K | 53.1 | Supported | 44.41–61.68 |
| 87 | [MiMo-V2-Pro](/models/mimo-v2-pro) | Xiaomi | Proprietary | 1M | 53 | Estimated | 43.25–62.70 |
| 88 | [MiMo-V2.5-Pro](/models/mimo-v2-5-pro) | Xiaomi | Proprietary | 1M | 53 | Supported | 35.98–69.97 |
| 89 | [Inkling](/models/inkling) | Thinking Machines Lab | Open Weight | 1M | 52.6 | Supported | 42.27–62.94 |
| 90 | [Hy3](/models/hy3) | Tencent | Open Weight | 256K | 52.5 | Estimated | 42.74–62.19 |
| 91 | [Hy3 Preview](/models/hy3-preview) | Tencent | Open Weight | 256K | 52.2 | Estimated | 37.86–66.56 |
| 92 | [Grok 4](/models/grok-4) | xAI | Proprietary | 128K | 51.8 | Supported | 42.00–61.54 |
| 93 | [GPT-5.1-Codex](/models/gpt-5-1-codex) | OpenAI | Proprietary | 400K | 50.4 | Estimated | 36.03–64.72 |
| 94 | [Apodex 1.1](/models/apodex-1-1) | Apodex | Proprietary | N/A | 50.2 | Estimated | 35.86–64.56 |
| 95 | [DeepSeek V3.2](/models/deepseek-v3-2) | DeepSeek | Open Weight | 128K | 50.2 | Supported | 36.76–63.65 |
| 96 | [MiMo-V2-Omni](/models/mimo-v2-omni) | Xiaomi | Proprietary | 262K | 50 | Estimated | 40.27–59.72 |
| 97 | [MiniMax M2.5](/models/minimax-m2-5) | MiniMax | Proprietary | 128K | 50 | Supported | 40.75–59.15 |
| 98 | [Gemini 3.5 Flash-Lite](/models/gemini-3-5-flash-lite) | Google | Proprietary | 1M | 49.6 | Supported | 36.97–62.28 |
| 99 | [GLM-5V-Turbo](/models/glm-5v-turbo) | Z.AI | Proprietary | 200K | 49.6 | Estimated | 39.88–59.33 |
| 100 | [Kimi K2.7 Code](/models/kimi-k2-7-code) | Moonshot AI | Open Weight | 256K | 49.5 | Supported | 42.30–56.74 |
| 101 | [Nemotron 3.5 Super VL 120B A12B](/models/nemotron-3-5-super-vl-120b-a12b) | NVIDIA | Open Weight | 1M | 49.4 | Estimated | 35.04–63.74 |
| 102 | [Gemini 2.5 Pro](/models/gemini-2-5-pro) | Google | Proprietary | 1M | 49.1 | Supported | 34.57–63.71 |
| 103 | [GLM-4.7](/models/glm-4-7) | Z.AI | Open Weight | 200K | 48.9 | Supported | 36.71–61.00 |
| 104 | [Qwen3.5 Plus](/models/qwen3-5-plus) | Alibaba | Proprietary | 1M | 48.9 | Estimated | 34.50–63.19 |
| 105 | [Qwen3.6-27B](/models/qwen3-6-27b) | Alibaba | Open Weight | 262K | 48.6 | Estimated | 34.24–62.94 |
| 106 | [Quasar 438B](/models/quasar-438b) | Multiverse Computing | Proprietary | 1M | 48.5 | Estimated | 34.12–62.81 |
| 107 | [Qwen3.5-27B](/models/qwen3-5-27b) | Alibaba | Open Weight | 262K | 48.4 | Estimated | 38.68–58.13 |
| 108 | [GPT-5 mini](/models/gpt-5-mini) | OpenAI | Proprietary | 128K | 48.4 | Supported | 37.02–59.77 |
| 109 | [MiniMax M2.7](/models/minimax-m2-7) | MiniMax | Open Weight | 200K | 48.2 | Supported | 36.79–59.55 |
| 110 | [Claude Sonnet 4.5](/models/claude-sonnet-4-5) | Anthropic | Proprietary | 200K | 48 | Estimated | 38.24–57.68 |
| 111 | [Gemini 3.1 Flash-Lite](/models/gemini-3-1-flash-lite) | Google | Proprietary | 1M | 47 | Estimated | 37.24–56.69 |
| 112 | [Qwen3.7 Flash](/models/qwen3-7-flash) | Alibaba | Proprietary | 1M | 46.9 | Estimated | 32.58–61.28 |
| 113 | [GPT-5.4 nano](/models/gpt-5-4-nano) | OpenAI | Proprietary | 400K | 46.9 | Supported | 36.22–57.51 |
| 114 | [Ling 3.0 Flash VL](/models/ling-3-0-flash-vl) | InclusionAI | Open Weight | 262K | 46.7 | Estimated | 32.37–61.07 |
| 115 | [Nemotron 3 Ultra](/models/nemotron-3-ultra) | NVIDIA | Open Weight | 1M | 46.6 | Estimated | 32.24–60.93 |
| 116 | [GPT-5 (medium)](/models/gpt-5-medium) | OpenAI | Proprietary | 128K | 46 | Estimated | 36.26–55.71 |
| 117 | [Grok 4.1 Fast (Reasoning)](/models/grok-4-1-fast-reasoning) | xAI | Proprietary | 2M | 45.8 | Estimated | 36.11–55.55 |
| 118 | [Qwen3.5-35B-A3B](/models/qwen3-5-35b-a3b) | Alibaba | Open Weight | 262K | 45.8 | Supported | 34.82–56.79 |
| 119 | [Gemma 4 26B A4B](/models/gemma-4-26b-a4b) | Google | Open Weight | 256K | 45.7 | Supported | 30.94–60.36 |
| 120 | [o3](/models/o3) | OpenAI | Proprietary | 200K | 45.4 | Supported | 39.20–51.68 |
| 121 | [o3-pro](/models/o3-pro) | OpenAI | Proprietary | 200K | 45.4 | Supported | 39.47–51.29 |
| 122 | [Nemotron 3.5 Bolt 74B-A9B](/models/nemotron-3-5-bolt-74b-a9b) | NVIDIA | Open Weight | 256K | 45.2 | Estimated | 30.86–59.56 |
| 123 | [Ling 3.0 Flash](/models/ling-3-0-flash) | InclusionAI | Open Weight | 262K | 45.1 | Estimated | 30.76–59.46 |
| 124 | [Grok 3 Mini](/models/grok-3-mini) | xAI | Proprietary | 128K | 44.2 | Estimated | 34.47–53.92 |
| 125 | [Step 3.7 Flash](/models/step-3-7-flash) | StepFun | Open Weight | 256K | 43.7 | Estimated | 29.30–58.00 |
| 126 | [Qwen3.6-35B-A3B](/models/qwen3-6-35b-a3b) | Alibaba | Open Weight | 262K | 43.5 | Estimated | 29.16–57.86 |
| 127 | [Grok 4 Fast (Reasoning)](/models/grok-4-fast-reasoning) | xAI | Proprietary | 2M | 43.5 | Estimated | 33.76–53.20 |
| 128 | [Qwen3.5 Flash](/models/qwen3-5-flash) | Alibaba | Proprietary | 1M | 42.8 | Estimated | 33.10–52.54 |
| 129 | [Gemini 2.5 Flash](/models/gemini-2-5-flash) | Google | Proprietary | 1M | 42.2 | Supported | 26.21–58.09 |
| 130 | [Step 3.5 Flash](/models/step-3-5-flash) | StepFun | Open Weight | 256K | 42.1 | Estimated | 32.39–51.84 |
| 131 | [Qwen3.5-122B-A10B](/models/qwen3-5-122b-a10b) | Alibaba | Open Weight | 262K | 41.7 | Estimated | 32.02–51.46 |
| 132 | [Muse Glimmer 30B](/models/muse-glimmer-30b) | Meta | Open Weight | 131K | 41.2 | Estimated | 26.84–55.54 |
| 133 | [Claude Haiku 4.5](/models/claude-haiku-4-5) | Anthropic | Proprietary | 200K | 41.2 | Estimated | 31.44–50.88 |
| 134 | [DeepSeek-R1](/models/deepseek-r1) | DeepSeek | Open Weight | 128K | 41.1 | Supported | 26.85–55.32 |
| 135 | [MiMo-V2-Flash](/models/mimo-v2-flash) | Xiaomi | Open Weight | 256K | 41 | Estimated | 31.28–50.72 |
| 136 | [Grok 3 [Beta]](/models/grok-3-beta) | xAI | Proprietary | 128K | 41 | Estimated | 31.25–50.70 |
| 137 | [DeepSeek V3.1](/models/deepseek-v3-1) | DeepSeek | Open Weight | 128K | 40.8 | Supported | 24.25–57.31 |
| 138 | [Claude 4.1 Opus](/models/claude-4-1-opus) | Anthropic | Proprietary | 200K | 40.1 | Estimated | 30.33–49.77 |
| 139 | [GLM-4.6](/models/glm-4-6) | Z.AI | Open Weight | 200K | 40 | Estimated | 30.29–49.74 |
| 140 | [Gemma 4 31B](/models/gemma-4-31b) | Google | Open Weight | 256K | 39.9 | Estimated | 30.16–49.61 |
| 141 | [Qwen3 Max](/models/qwen3-max) | Alibaba | Proprietary | 1M | 39.6 | Estimated | 25.22–53.92 |
| 142 | [o3-mini](/models/o3-mini) | OpenAI | Proprietary | 200K | 39.6 | Supported | 27.64–51.47 |
| 143 | [Claude 4.1 Opus Thinking](/models/claude-4-1-opus-thinking) | Anthropic | Proprietary | 200K | 39.4 | Estimated | 29.65–49.09 |
| 144 | [DeepSeek V3.1 (Reasoning)](/models/deepseek-v3-1-reasoning) | DeepSeek | Open Weight | 128K | 39.2 | Estimated | 29.46–48.90 |
| 145 | [GLM-4.7-Flash](/models/glm-4-7-flash) | Z.AI | Open Weight | 200K | 39.1 | Estimated | 29.41–48.86 |
| 146 | [o1](/models/o1) | OpenAI | Proprietary | 200K | 38.4 | Estimated | 24.03–52.73 |
| 147 | [Claude 4 Sonnet](/models/claude-4-sonnet) | Anthropic | Proprietary | 200K | 38.1 | Estimated | 28.36–47.80 |
| 148 | [LFM2.5-24B-A2B](/models/lfm2-5-24b-a2b) | LiquidAI | Open Weight | 131K | 37.9 | Estimated | 23.55–52.25 |
| 149 | [GPT-4.1](/models/gpt-4-1) | OpenAI | Proprietary | 1M | 37.6 | Supported | 21.73–53.46 |
| 150 | [K-Exaone](/models/k-exaone) | LG AI Research | Proprietary | 256K | 37.5 | Estimated | 23.15–51.85 |
| 151 | [GPT-5 nano](/models/gpt-5-nano) | OpenAI | Proprietary | 400K | 37.4 | Supported | 25.17–49.56 |
| 152 | [GPT-OSS 120B](/models/gpt-oss-120b) | OpenAI | Open Weight | 128K | 37.3 | Supported | 23.58–51.06 |
| 153 | [DeepSeek V3 0324](/models/deepseek-v3-0324) | DeepSeek | Open Weight | 128K | 36.5 | Supported | 20.08–52.91 |
| 154 | [Grok 4.1 Fast](/models/grok-4-1-fast) | xAI | Proprietary | 1M | 36.5 | Estimated | 26.73–46.18 |
| 155 | [Gemma 4 12B](/models/gemma-4-12b) | Google | Open Weight | 256K | 36.4 | Estimated | 22.00–50.70 |
| 156 | [Command A+](/models/command-a-plus) | Cohere | Open Weight | 128K | 36.1 | Estimated | 21.79–50.49 |
| 157 | [o1-preview](/models/o1-preview) | OpenAI | Proprietary | 200K | 36 | Supported | 19.52–52.45 |
| 158 | [o1-pro](/models/o1-pro) | OpenAI | Proprietary | 200K | 35.7 | Estimated | 21.37–50.07 |
| 159 | [GLM-4.5-Air](/models/glm-4-5-air) | Z.AI | Proprietary | 128K | 34.9 | Estimated | 25.21–44.66 |
| 160 | [Mistral Small 4](/models/mistral-small-4) | Mistral | Open Weight | 256K | 34.7 | Estimated | 24.98–44.43 |
| 161 | [Trinity-Large-Thinking](/models/trinity-large-thinking) | Arcee AI | Open Weight | 512K | 34.5 | Estimated | 24.78–44.23 |
| 162 | [ZAYA1-8B](/models/zaya1-8b) | Zyphra | Open Weight | 131K | 34.2 | Estimated | 19.86–48.56 |
| 163 | [Grok Code Fast 1](/models/grok-code-fast-1) | xAI | Proprietary | 256K | 34.1 | Estimated | 24.37–43.81 |
| 164 | [Ornith-1.5-35B-A3B](/models/ornith-1-5-35b-a3b) | Ornith AI | Open Weight | 262K | 34 | Estimated | 19.63–48.33 |
| 165 | [Mercury 2.5](/models/mercury-2-5) | Inception | Proprietary | 260K | 33.8 | Estimated | 19.42–48.12 |
| 166 | [Mistral Medium 3.5 128B](/models/mistral-medium-3-5-128b) | Mistral | Open Weight | 256K | 33.6 | Estimated | 19.27–47.97 |
| 167 | [Mercury 2](/models/mercury-2) | Inception | Proprietary | 128K | 33.5 | Estimated | 23.80–43.24 |
| 168 | [Kimi K2](/models/kimi-k2) | Moonshot AI | Proprietary | 128K | 33.5 | Estimated | 23.76–43.20 |
| 169 | [Mistral Large 3](/models/mistral-large-3) | Mistral | Proprietary | 128K | 33.5 | Estimated | 23.76–43.20 |
| 170 | [GPT-OSS 20B](/models/gpt-oss-20b) | OpenAI | Open Weight | 128K | 33.3 | Supported | 20.00–46.64 |
| 171 | [Granite 4.2 8B](/models/granite-4-2-8b) | IBM | Open Weight | 128K | 33.1 | Estimated | 23.36–42.81 |
| 172 | [Ling 2.6 Flash](/models/ling-2-6-flash) | InclusionAI | Open Weight | 262K | 32.4 | Estimated | 18.08–46.78 |
| 173 | [Nemotron 3 Nano Omni 30B A3B](/models/nemotron-3-nano-omni-30b-a3b) | NVIDIA | Open Weight | 256K | 32.2 | Estimated | 17.87–46.57 |
| 174 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | Open Weight | 128K | 31.9 | Supported | 15.52–48.28 |
| 175 | [Sarvam 105B](/models/sarvam-105b) | Sarvam | Open Weight | 128K | 31.6 | Estimated | 17.29–45.99 |
| 176 | [Ornith-1.5-9B](/models/ornith-1-5-9b) | Ornith AI | Open Weight | 262K | 31.3 | Estimated | 16.98–45.67 |
| 177 | [LFM2.5-2.6B](/models/lfm2-5-2-6b) | LiquidAI | Open Weight | 128K | 31.2 | Estimated | 16.85–45.55 |
| 178 | [Gemma 4 E4B](/models/gemma-4-e4b) | Google | Open Weight | 128K | 31 | Estimated | 16.67–45.37 |
| 179 | [DeepSeek R1 Distill Qwen 32B](/models/deepseek-r1-distill-qwen-32b) | DeepSeek | Open Weight | 128K | 30.8 | Estimated | 16.46–45.16 |
| 180 | [LFM2.5-8B-A1B](/models/lfm2-5-8b-a1b) | LiquidAI | Open Weight | 128K | 30.1 | Estimated | 15.77–44.47 |
| 181 | [Solar Pro 3](/models/solar-pro-3) | Upstage | Proprietary | 128K | 30 | Estimated | 15.63–44.33 |
| 182 | [Gemma 4 E2B](/models/gemma-4-e2b) | Google | Open Weight | 128K | 29.9 | Estimated | 15.59–44.29 |
| 183 | [GPT-4o](/models/gpt-4o) | OpenAI | Proprietary | 128K | 29.9 | Estimated | 20.20–39.64 |
| 184 | [GPT-4.1 mini](/models/gpt-4-1-mini) | OpenAI | Proprietary | 1M | 29.3 | Estimated | 14.92–43.62 |
| 185 | [Solar Pro 2](/models/solar-pro-2) | Upstage | Proprietary | 128K | 29 | Estimated | 14.68–43.37 |
| 186 | [Granite-4.0-H-1B](/models/granite-4-0-h-1b) | IBM | Open Weight | 128K | 28.8 | Estimated | 14.46–43.16 |
| 187 | [Mistral Medium 3](/models/mistral-medium-3) | Mistral | Proprietary | 128K | 28.8 | Estimated | 14.45–43.15 |
| 188 | [Qwen2.5-72B](/models/qwen2-5-72b) | Alibaba | Open Weight | 128K | 28.7 | Estimated | 19.02–38.46 |
| 189 | [Gemma 3 27B](/models/gemma-3-27b) | Google | Open Weight | 32K | 28.7 | Supported | 9.88–47.51 |
| 190 | [Sarvam 30B](/models/sarvam-30b) | Sarvam | Open Weight | 64K | 28.6 | Estimated | 14.26–42.96 |
| 191 | [Claude 3 Opus](/models/claude-3-opus) | Anthropic | Proprietary | 200K | 28.5 | Supported | 13.24–43.74 |
| 192 | [Granite-4.0-350M](/models/granite-4-0-350m) | IBM | Open Weight | 32K | 28.5 | Estimated | 14.10–42.80 |
| 193 | [Granite-4.0-H-350M](/models/granite-4-0-h-350m) | IBM | Open Weight | 32K | 28.5 | Estimated | 14.10–42.80 |
| 194 | [Celeris-1](/models/celeris-1) | Celeris | Proprietary | 128K | 28.4 | Estimated | 14.06–42.76 |
| 195 | [Exaone 4.0 32B](/models/exaone-4-0-32b) | LG AI Research | Open Weight | 128K | 28.4 | Estimated | 14.04–42.74 |
| 196 | [Mistral Large 2](/models/mistral-large-2) | Mistral | Proprietary | 128K | 28.4 | Estimated | 18.63–38.07 |
| 197 | [Llama 3.1 405B](/models/llama-3-1-405b) | Meta | Open Weight | 128K | 28.2 | Estimated | 18.48–37.93 |
| 198 | [Llama 4 Scout](/models/llama-4-scout) | Meta | Open Weight | 10M | 28.2 | Supported | 12.92–43.47 |
| 199 | [Qwen3-Omni-30B-A3B-Instruct](/models/qwen3-omni-30b-a3b-instruct) | Alibaba | Open Weight | N/A | 28.1 | Estimated | 13.73–42.43 |
| 200 | [Gemini 1.5 Pro](/models/gemini-1-5-pro) | Google | Proprietary | 2M | 27.8 | Estimated | 18.05–37.49 |
| 201 | [Claude 3.5 Sonnet](/models/claude-3-5-sonnet) | Anthropic | Proprietary | 200K | 27.7 | Estimated | 18.00–37.45 |
| 202 | [MiniMax M1 80k](/models/minimax-m1-80k) | MiniMax | Proprietary | 80K | 27.6 | Estimated | 17.86–37.31 |
| 203 | [Exaone 4.0 1.2B](/models/exaone-4-0-1-2b) | LG AI Research | Open Weight | 128K | 27.4 | Estimated | 13.00–41.70 |
| 204 | [GPT-4o mini](/models/gpt-4o-mini) | OpenAI | Proprietary | 128K | 26.8 | Supported | 10.94–42.65 |
| 205 | [Qwen2.5 Coder 32B Instruct](/models/qwen2-5-coder-32b-instruct) | Alibaba | Open Weight | 128K | 26.6 | Estimated | 16.89–36.33 |
| 206 | [Nemotron 3.5 Lightning 30B A3B NVFP4](/models/nemotron-3-5-lightning-30b-a3b-nvfp4) | NVIDIA | Open Weight | 1M | 25.5 | Estimated | 11.15–39.85 |
| 207 | [Phi-4](/models/phi-4) | Microsoft | Open Weight | 16K | 25.1 | Supported | 13.29–36.86 |
| 208 | [Ministral 3 14B](/models/ministral-3-14b) | Mistral | Open Weight | 128K | 24.9 | Estimated | 15.22–34.66 |
| 209 | [GPT-4.1 nano](/models/gpt-4-1-nano) | OpenAI | Proprietary | 1M | 24.5 | Estimated | 10.18–38.88 |
| 210 | [Nova Pro](/models/nova-pro) | Amazon | Proprietary | 128K | 21.7 | Estimated | 11.95–31.40 |
| 211 | [Llama 4 Maverick](/models/llama-4-maverick) | Meta | Open Weight | 1M | 21.6 | Supported | 16.95–26.27 |
| 212 | [GPT-4 Turbo](/models/gpt-4-turbo) | OpenAI | Proprietary | 128K | 21.5 | Supported | 14.93–28.15 |
| 213 | [Llama 3 70B](/models/llama-3-70b) | Meta | Open Weight | 128K | 21.3 | Estimated | 11.59–31.04 |
| 214 | [LFM2-24B-A2B](/models/lfm2-24b-a2b) | LiquidAI | Proprietary | 32K | 20.7 | Estimated | 10.94–30.39 |
| 215 | [Ministral 3 8B](/models/ministral-3-8b) | Mistral | Open Weight | 128K | 20 | Estimated | 10.32–29.77 |
| 216 | [LFM2.5-1.2B-Thinking](/models/lfm2-5-1-2b-thinking) | LiquidAI | Proprietary | 32K | 19.4 | Estimated | 9.71–29.16 |
| 217 | [LFM2.5-1.2B-Instruct](/models/lfm2-5-1-2b-instruct) | LiquidAI | Proprietary | 32K | 19.2 | Estimated | 9.46–28.91 |
| 218 | [Ministral 3 3B](/models/ministral-3-3b) | Mistral | Open Weight | 128K | 18.9 | Estimated | 9.16–28.60 |
| 219 | [Mixtral 8x22B Instruct v0.1](/models/mixtral-8x22b-instruct-v0-1) | Mistral | Open Weight | 64K | 18.2 | Estimated | 8.44–27.89 |
| 220 | [Claude 3 Haiku](/models/claude-3-haiku) | Anthropic | Proprietary | 200K | 14.2 | Supported | 5.82–22.56 |
| 221 | [Gemini 1.0 Pro](/models/gemini-1-0-pro) | Google | Proprietary | 32K | 11.7 | Supported | 1.08–22.22 |
| 222 | [Nemotron-4 15B](/models/nemotron-4-15b) | NVIDIA | Open Weight | 32K | 9.9 | Estimated | 0.21–19.66 |
| 223 | [Mistral 7B v0.3](/models/mistral-7b-v0-3) | Mistral | Open Weight | 32K | 7.2 | Estimated | 0.00–16.93 |
| 224 | [MiniCPM5-1B](/models/minicpm5-1b) | OpenBMB | Open Weight | 131K | 6.3 | Estimated | 0.00–20.63 |
| — | [Mistral Large 4](/models/mistral-large-4) | Mistral | Pending | 1M | — | Tracked | — |

## Key Takeaways

- Top model: [Claude Opus 5.5](/models/claude-opus-5-5) with a score of 86.6 and Supported evidence
- Best open-weight option: [MiMo-V2.6-Pro](/models/mimo-v2-6-pro) at #12
- Models included: 225 (224 ranked, 1 listed without a rank)

## Compare the Leaders

- [Claude Opus 5.5 vs GPT-6.1 Sol](/compare/claude-opus-5-5-vs-gpt-6-1-sol)