| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 21 | Gemini 3 Pro gemini-3-pro-preview multimodalvisionmulti-input reasoning | Google | 61.0 overall | 70.3 | 0.0 | 58.0 | 52.3 | 0.0 | N/A |
| 22 | GPT-5.1 Codex High gpt-5.1-codex-high multimodalvisionmulti-input reasoning | OpenAI | 61.0 overall | 61.0 | 0.0 | 0.0 | 0.0 | 0.0 | |
| 23 | Qwen3.7 Max qwen3.7-max multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 60.8 overall | 64.7 | 62.8 | 49.6 | 74.2 | 43.9 | $1.25 in / $3.75 out |
| 24 | Nova 2 Sonic nova-2-sonic multimodalvisionmulti-input reasoning | Amazon | 60.7 overall | 0.0 | 62.8 | 0.0 | 0.0 | 57.3 | $0.33 in / $2.75 out |
| 25 | GPT-5.5 Pro gpt-5.5-pro multimodalvisionmulti-input reasoning | OpenAI | 60.1 overall | 61.1 | 0.0 | 69.2 | 48.8 | 0.0 | N/A |
| 26 | DeepSeek-V4-Pro-Max deepseek-v4-pro-max codeprogrammingtool use | DeepSeek | 59.9 overall | 64.9 | 84.8 | 51.0 | 52.0 | 45.1 | |
| 27 | Qwen3.7-Plus qwen3.7-plus multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 59.6 overall | 62.9 | 62.8 | 48.1 | 61.1 | 69.5 | $0.32 in / $1.28 out |
| 28 | Gemini 3.6 Flash gemini-3.6-flash multimodalvisionmulti-input reasoning | Google | 59.0 overall | 57.8 | 84.8 | 0.0 | 52.6 | 34.8 | |
| 29 | Kimi K2-Thinking-0905 kimi-k2-thinking-0905 codeprogrammingtool use | Moonshot AI | 58.8 overall | 66.2 | 0.0 | 50.8 | 58.0 | 0.0 | |
| 30 | Grok 4.3 grok-4.3 textinference | xAI | 58.8 overall | 0.0 | 62.8 | 0.0 | 0.0 | 52.4 | $1.25 in / $2.5 out |
| 31 | Grok 4.5 grok-4.5 multimodalvisionmulti-input reasoning | xAI | 58.7 overall | 69.3 | 38.2 | 0.0 | 70.8 | 35.6 | $2 in / $6 out |
| 32 | Gemini 3.1 Flash-Lite gemini-3.1-flash-lite-preview multimodalvisionmulti-input reasoning | Google | 58.3 overall | 52.6 | 62.8 | 0.0 | 0.0 | 67.1 | |
| 33 | Gemini 3.1 Pro gemini-3.1-pro-preview multimodalvisionmulti-input reasoning | Google | 57.9 overall | 72.3 | 58.3 | 52.6 | 60.9 | 22.7 | |
| 34 | MiMo-V2.5-Pro mimo-v2.5-pro codeprogrammingtool use | Xiaomi | 57.7 overall | 36.2 | 84.8 | 0.0 | 56.3 | 78.0 | $0.435 in / $0.87 out |
| 35 | GPT-5.1 Thinking gpt-5.1-thinking-2025-11-12 multimodalvisionmulti-input reasoning | OpenAI | 57.6 overall | 62.4 | 0.0 | 0.0 | 51.4 | 0.0 | |
| 36 | Claude Opus 4.1 claude-opus-4-1-20250805 multimodalvisionmulti-input reasoning | Anthropic | 57.3 overall | 46.0 | 0.0 | 67.4 | 60.8 | 0.0 | |
| 37 | Muse Spark muse-spark multimodalvisionmulti-input reasoning | Meta | 57.0 overall | 67.1 | 0.0 | 64.4 | 36.2 | 0.0 | N/A |
| 38 | ERNIE 5.0 ernie-5.0 multimodalvisionmulti-input reasoning | Baidu | 56.8 overall | 56.8 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 39 | DeepSeek-V4-Flash-Max deepseek-v4-flash-max codeprogrammingtool use | DeepSeek | 56.7 overall | 56.2 | 84.8 | 35.3 | 41.4 | 98.8 | |
| 40 | Claude Opus 4.7 claude-opus-4-7 multimodalvisionmulti-input reasoning | Anthropic | 56.5 overall | 75.1 | 28.3 | 54.2 | 77.9 | 8.0 |
21Google
Gemini 3 Pro
61.0
N/A
22
GPT-5.1 Codex High
OpenAI
61.0
N/A
23
Qwen3.7 Max
Alibaba Cloud / Qwen Team
60.8
$1.25 in / $3.75 out
24