| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 41 | Qwen3.5-122B-A10B qwen3.5-122b-a10b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 44.6 Agentic | 60.5 | 0.0 | 44.6 | 38.3 | 0.0 | N/A |
| 42 | GLM-5.2 glm-5.2 codeprogrammingtool use | Zhipu AI | 44.1 Agentic | 68.7 | 84.8 | 44.1 | 59.9 | 51.2 | $0.95 in / $3 out |
| 43 | MiniMax M2.5 minimax-m2.5 codeprogrammingtool use | MiniMax | 43.6 Agentic | 0.0 | 68.9 | 43.6 | 50.4 | 72.9 | $0.3 in / $1.2 out |
| 44 | Qwen3-Next-80B-A3B-Thinking qwen3-next-80b-a3b-thinking textinference | Alibaba Cloud / Qwen Team | 41.7 Agentic | 42.3 | 0.0 | 41.7 | 0.0 | 0.0 | N/A |
| 45 | Qwen3.5-27B qwen3.5-27b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 41.3 Agentic | 57.8 | 32.9 | 41.3 | 39.1 | 61.0 | $0.3 in / $2.4 out |
| 46 | Claude Sonnet 4.6 claude-sonnet-4-6 multimodalvisionmulti-input reasoning | Anthropic | 41.0 Agentic | 62.3 | 12.6 | 41.0 | 65.6 | 12.0 | |
| 47 | Kimi K2.5 kimi-k2.5 multimodalvisionmulti-input reasoning | Moonshot AI | 41.0 Agentic | 63.7 | 0.0 | 41.0 | 41.8 | 0.0 | N/A |
| 48 | Hy3 hy3 codeprogrammingtool use | Tencent | 40.4 Agentic | 65.9 | 0.0 | 40.4 | 59.2 | 0.0 | N/A |
| 49 | MiniMax M2 minimax-m2 codeprogrammingtool use | MiniMax | 40.2 Agentic | 29.8 | 62.8 | 40.2 | 40.0 | 73.2 | $0.3 in / $1.2 out |
| 50 | Claude 3.5 Sonnet claude-3-5-sonnet-20241022 multimodalvisionmulti-input reasoning | Anthropic | 38.7 Agentic | 32.0 | 0.0 | 38.7 | 11.1 | 0.0 | |
| 51 | Qwen3.5-35B-A3B qwen3.5-35b-a3b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 38.3 Agentic | 53.4 | 0.0 | 38.3 | 30.2 | 0.0 | N/A |
| 52 | MiniMax M3 minimax-m3 multimodalvisionmulti-input reasoning | MiniMax | 37.5 Agentic | 49.6 | 62.8 | 37.5 | 68.3 | 73.2 | $0.3 in / $1.2 out |
| 53 | GPT-5.2 gpt-5.2-2025-12-11 multimodalvisionmulti-input reasoning | OpenAI | 37.0 Agentic | 70.9 | 62.8 | 37.0 | 66.3 | 31.5 | |
| 54 | Step-3.5-Flash step-3.5-flash codeprogrammingtool use | StepFun | 36.5 Agentic | 62.8 | 59.8 | 36.5 | 48.6 | 93.9 | $0.1 in / $0.4 out |
| 55 | Qwen3 VL 235B A22B Thinking qwen3-vl-235b-a22b-thinking multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 36.2 Agentic | 35.2 | 0.0 | 36.2 | 0.0 | 0.0 | |
| 56 | o4-mini o4-mini multimodalvisionmulti-input reasoning | OpenAI | 36.1 Agentic | 46.2 | 0.0 | 36.1 | 28.7 | 0.0 | N/A |
| 57 | GLM-4.5 glm-4.5 codeprogrammingtool use | Zhipu AI | 36.0 Agentic | 31.5 | 0.0 | 36.0 | 36.1 | 0.0 | N/A |
| 58 | GPT-4.5 gpt-4.5 multimodalvisionmulti-input reasoning | OpenAI | 35.8 Agentic | 41.1 | 0.0 | 35.8 | 5.2 | 0.0 | N/A |
| 59 | GLM-4.6 glm-4.6 multimodalvisionmulti-input reasoning | Zhipu AI | 35.4 Agentic | 44.5 | 0.0 | 35.4 | 43.2 | 0.0 | N/A |
| 60 | DeepSeek-V4-Flash-Max deepseek-v4-flash-max codeprogrammingtool use | DeepSeek | 35.3 Agentic | 56.2 | 84.8 | 35.3 | 41.4 | 98.8 |
41
Qwen3.5-122B-A10B
Alibaba Cloud / Qwen Team
44.6
N/A
42
GLM-5.2
Zhipu AI
44.1
$0.95 in / $3 out
43
MiniMax M2.5
MiniMax
43.6
$0.3 in / $1.2 out
44