| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 61 | o4-mini o4-mini multimodalvisionmulti-input reasoning | OpenAI | 48.8 Benchmarks | 48.8 | 70.7 | 38.2 | 32.7 | 41.9 | $1.1 in / $4.4 out |
| 62 | Claude Opus 4.1 claude-opus-4-1-20250805 multimodalvisionmulti-input reasoning | Anthropic | 48.1 Benchmarks | 48.1 | 30.1 | 66.8 | 62.9 | 7.0 | |
| 63 | o1-pro o1-pro multimodalvisionmulti-input reasoning | OpenAI | 47.5 Benchmarks | 47.5 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 64 | Step3-VL-10B step3-vl-10b multimodalvisionmulti-input reasoning | StepFun | 47.4 Benchmarks | 47.4 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 65 | GLM-4.6 glm-4.6 multimodalvisionmulti-input reasoning | Zhipu AI | 47.0 Benchmarks | 47.0 | 34.9 | 37.7 | 46.1 | 42.8 | $0.55 in / $2.19 out |
| 66 | Qwen3-235B-A22B-Thinking-2507 qwen3-235b-a22b-thinking-2507 textinference | Alibaba Cloud / Qwen Team | 46.9 Benchmarks | 46.9 | 66.8 | 26.8 | 0.0 | 39.4 | $0.3 in / $3 out |
| 67 | Gemini 2.0 Flash Thinking gemini-2.0-flash-thinking multimodalvisionmulti-input reasoning | Google | 46.7 Benchmarks | 46.7 | 0.0 | 0.0 | 0.0 | 0.0 | |
| 68 | Sarvam-30B sarvam-30b codeprogrammingtool use | Sarvam AI | 46.5 Benchmarks | 46.5 | 0.0 | 8.5 | 5.3 | 0.0 | N/A |
| 69 | o3 o3-2025-04-16 multimodalvisionmulti-input reasoning | OpenAI | 46.2 Benchmarks | 46.2 | 38.4 | 20.5 | 30.7 | 27.7 | $2 in / $8 out |
| 70 | GPT-5.4 nano gpt-5.4-nano multimodalvisionmulti-input reasoning | OpenAI | 46.1 Benchmarks | 46.1 | 77.4 | 11.0 | 11.2 | 57.2 | |
| 71 | Nemotron 3 Nano (30B A3B) nemotron-3-nano-30b-a3b codeprogrammingtool use | NVIDIA | 45.8 Benchmarks | 45.8 | 66.8 | 3.3 | 4.4 | 90.8 | $0.06 in / $0.24 out |
| 72 | GPT OSS 120B High gpt-oss-120b-high multimodalvisionmulti-input reasoning | OpenAI | 44.9 Benchmarks | 44.9 | 57.3 | 0.0 | 0.0 | 73.2 | |
| 73 | Qwen3-Next-80B-A3B-Thinking qwen3-next-80b-a3b-thinking textinference | Alibaba Cloud / Qwen Team | 44.9 Benchmarks | 44.9 | 6.1 | 41.7 | 0.0 | 51.9 | $0.15 in / $1.5 out |
| 74 | Gemini 2.5 Pro gemini-2.5-pro multimodalvisionmulti-input reasoning | Google | 44.6 Benchmarks | 44.6 | 63.2 | 0.0 | 25.6 | 27.9 | |
| 75 | Mercury 2 mercury-2 codeprogrammingtool use | Inception | 44.6 Benchmarks | 44.6 | 72.5 | 0.0 | 22.3 | 69.2 | $0.25 in / $0.75 out |
| 76 | Qwen3 VL 32B Thinking qwen3-vl-32b-thinking multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 44.6 Benchmarks | 44.6 | 0.0 | 34.6 | 0.0 | 0.0 | |
| 77 | Kimi K2 0905 kimi-k2-0905 textinference | Moonshot AI | 44.4 Benchmarks | 44.4 | 66.8 | 0.0 | 0.0 | 40.0 | $0.6 in / $2.5 out |
| 78 | Claude 3.7 Sonnet claude-3-7-sonnet-20250219 multimodalvisionmulti-input reasoning | Anthropic | 43.7 Benchmarks | 43.7 | 30.1 | 49.0 | 40.1 | 13.2 | |
| 79 | Gemma 4 26B-A4B gemma-4-26b-a4b-it multimodalvisionmulti-input reasoning | Google | 43.7 Benchmarks | 43.7 | 66.8 | 0.0 | 0.0 | 77.8 | |
| 80 | K-EXAONE-236B-A23B k-exaone-236b-a23b multimodalvisionmulti-input reasoning | LG AI Research | 43.4 Benchmarks | 43.4 | 24.2 | 0.0 | 0.0 | 49.1 | $0.6 in / $1 out |
61
o4-mini
OpenAI
48.8
$1.1 in / $4.4 out
62
Claude Opus 4.1
Anthropic
48.1
$15 in / $75 out
63
o1-pro
OpenAI
47.5
N/A
64