| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 21 | Gemini 3 Flash gemini-3-flash-preview multimodalvisionmulti-input reasoning | Google | 62.3 overall | 71.3 | 84.9 | 42.5 | 66.6 | 38.9 | $0.5 in / $3 out |
| 22 | Kimi K2-Thinking-0905 kimi-k2-thinking-0905 codeprogrammingtool use | Moonshot AI | 62.2 overall | 69.3 | 0.0 | 53.5 | 62.5 | 0.0 | |
| 23 | DeepSeek-V3.2 (Non-thinking) deepseek-chat textinference | DeepSeek | 62.2 overall | 0.0 | 57.3 | 0.0 | 0.0 | 70.1 | $0.28 in / $0.42 out |
| 24 | Kimi K2.6 kimi-k2.6 multimodalvisionmulti-input reasoning | Moonshot AI | 61.9 overall | 68.5 | 66.8 | 45.3 | 81.0 | 33.3 | $0.95 in / $4 out |
| 25 | Seed 2.0 Pro seed-2.0-pro multimodalvisionmulti-input reasoning | ByteDance | 61.9 overall | 68.2 | 0.0 | 54.7 | 61.8 | 0.0 | N/A |
| 26 | Qwen3.6 Plus qwen3.6-plus multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 61.7 overall | 71.9 | 0.0 | 49.3 | 62.2 | 0.0 | N/A |
| 27 | Muse Spark muse-spark multimodalvisionmulti-input reasoning | Meta | 61.0 overall | 71.0 | 0.0 | 67.3 | 41.3 | 0.0 | N/A |
| 28 | Claude Opus 4.6 claude-opus-4-6 multimodalvisionmulti-input reasoning | Anthropic | 60.9 overall | 79.5 | 42.8 | 60.7 | 73.3 | 10.6 | |
| 29 | Gemini 2.0 Flash gemini-2.0-flash multimodalvisionmulti-input reasoning | Google | 60.5 overall | 33.4 | 94.1 | 0.0 | 0.0 | 82.7 | |
| 30 | GPT-5.4 gpt-5.4 texttext-to-textlanguage | OpenAI | 60.3 overall | 76.3 | 51.1 | 63.8 | 62.1 | 18.2 | |
| 31 | GPT-5.1 gpt-5.1-2025-11-13 multimodalvisionmulti-input reasoning | OpenAI | 59.8 overall | 65.0 | 71.4 | 0.0 | 57.2 | 31.9 | |
| 32 | GPT-5.1 Instant gpt-5.1-instant-2025-11-12 multimodalvisionmulti-input reasoning | OpenAI | 59.8 overall | 65.0 | 71.4 | 0.0 | 57.2 | 31.9 | |
| 33 | ERNIE 5.0 ernie-5.0 multimodalvisionmulti-input reasoning | Baidu | 59.7 overall | 59.7 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 34 | MiniMax M2.5 minimax-m2.5 codeprogrammingtool use | MiniMax | 59.3 overall | 0.0 | 73.9 | 53.0 | 56.3 | 57.7 | $0.3 in / $1.2 out |
| 35 | Llama 4 Scout llama-4-scout multimodalvisionmulti-input reasoning | Meta | 58.8 overall | 29.2 | 93.0 | 0.0 | 0.0 | 87.2 | $0.08 in / $0.3 out |
| 36 | Ministral 3 (8B Reasoning 2512) ministral-8b-latest multimodalvisionmulti-input reasoning | Mistral AI | 58.6 overall | 31.8 | 84.8 | 0.0 | 0.0 | 92.1 | |
| 37 | Step-3.5-Flash step-3.5-flash codeprogrammingtool use | StepFun | 58.3 overall | 62.3 | 63.2 | 45.3 | 53.0 | 82.1 | $0.1 in / $0.4 out |
| 38 | Ministral 3 (14B Reasoning 2512) ministral-14b-latest multimodalvisionmulti-input reasoning | Mistral AI | 58.0 overall | 37.9 | 76.8 | 0.0 | 0.0 | 84.5 | |
| 39 | Gemma 4 26B-A4B gemma-4-26b-a4b-it multimodalvisionmulti-input reasoning | Google | 56.8 overall | 43.7 | 66.8 | 0.0 | 0.0 | 77.8 | |
| 40 | GPT-5.1 Medium gpt-5.1-medium-2025-11-12 multimodalvisionmulti-input reasoning | OpenAI | 56.6 overall | 63.6 | 61.6 | 0.0 | 0.0 | 29.0 |
21Google
Gemini 3 Flash
62.3
$0.5 in / $3 out
22
Kimi K2-Thinking-0905
Moonshot AI
62.2
N/A
23
DeepSeek-V3.2 (Non-thinking)
DeepSeek
62.2
$0.28 in / $0.42 out
24