| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 121 | Nemotron 3.5 Super VL (120B A12B) nemotron-3.5-super-vl-120b-a12b multimodalvisionmulti-input reasoning | NVIDIA | 46.9 overall | 52.2 | 41.8 | 48.2 | 35.3 | 62.9 | $0.2 in / $1 out |
| 122 | Qwen3-235B-A22B-Instruct-2507 qwen3-235b-a22b-instruct-2507 textinference | Alibaba Cloud / Qwen Team | 46.9 overall | 39.7 | 41.8 | 0.0 | 0.0 | 75.5 | $0.09 in / $0.55 out |
| 123 | Llama Guard 4 12B llama-guard-4-12b multimodalvisionmulti-input reasoning | Meta | 46.9 overall | 0.0 | 22.1 | 0.0 | 0.0 | 86.5 | $0.18 in / $0.18 out |
| 124 | Grok-3 grok-3 multimodalvisionmulti-input reasoning | xAI | 46.8 overall | 56.8 | 46.0 | 0.0 | 0.0 | 20.3 | $3 in / $15 out |
| 125 | Qwen3.5-122B-A10B qwen3.5-122b-a10b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 46.1 overall | 59.7 | 41.8 | 40.9 | 38.0 | 45.1 | $0.29 in / $2.4 out |
| 126 | GPT-5.5 Instant gpt-5.5-instant multimodalvisionmulti-input reasoning | OpenAI | 46.1 overall | 48.3 | 61.1 | 0.0 | 0.0 | 15.7 | |
| 127 | Claude Opus 4 claude-opus-4-20250514 multimodalvisionmulti-input reasoning | Anthropic | 46.1 overall | 35.8 | 0.0 | 57.4 | 46.7 | 0.0 | |
| 128 | Step3-VL-10B step3-vl-10b multimodalvisionmulti-input reasoning | StepFun | 46.0 overall | 46.0 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 129 | Inkling inkling multimodalvisionmulti-input reasoning | Thinking Machines Lab | 45.8 overall | 34.1 | 58.2 | 45.7 | 58.9 | 30.5 | $0.95 in / $4.05 out |
| 130 | Mercury 2 mercury-2 codeprogrammingtool use | Inception | 45.1 overall | 41.7 | 69.0 | 0.0 | 19.2 | 73.4 | $0.25 in / $0.75 out |
| 131 | Qwen3.6-27B qwen3.6-27b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 45.0 overall | 52.9 | 41.8 | 0.0 | 41.1 | 36.8 | $0.32 in / $3.2 out |
| 132 | K-EXAONE-236B-A23B k-exaone-236b-a23b multimodalvisionmulti-input reasoning | LG AI Research | 44.7 overall | 44.7 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 133 | GPT-5 mini gpt-5-mini-2025-08-07 multimodalvisionmulti-input reasoning | OpenAI | 44.7 overall | 38.2 | 79.9 | 0.0 | 22.2 | 56.3 | |
| 134 | Qwen3.5-27B qwen3.5-27b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 44.5 overall | 57.2 | 41.8 | 38.1 | 38.6 | 41.8 | $0.26 in / $2.6 out |
| 135 | Seed 2.0 Mini seed-2.0-mini multimodalvisionmulti-input reasoning | ByteDance | 44.4 overall | 39.3 | 30.9 | 0.0 | 0.0 | 80.2 | $0.1 in / $0.4 out |
| 136 | Qwen3-Coder 480B A35B Instruct qwen3-coder-480b-a35b-instruct codeprogrammingtool use | Alibaba Cloud / Qwen Team | 44.1 overall | 0.0 | 41.8 | 50.7 | 30.6 | 61.5 | |
| 137 | Claude Haiku 4.5 claude-haiku-4-5-20251001 multimodalvisionmulti-input reasoning | Anthropic | 44.0 overall | 30.1 | 51.8 | 49.8 | 54.1 | 34.2 | |
| 138 | GPT-5 gpt-5-2025-08-07 multimodalvisionmulti-input reasoning | OpenAI | 43.9 overall | 58.5 | 0.0 | 24.6 | 46.5 | 0.0 | N/A |
| 139 | Qwen3 VL 235B A22B Instruct qwen3-vl-235b-a22b-instruct multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 43.8 overall | 32.6 | 41.8 | 49.3 | 0.0 | 65.4 | |
| 140 | Llama 4 Scout llama-4-scout multimodalvisionmulti-input reasoning | Meta | 43.7 overall | 25.9 | 49.2 | 0.0 | 0.0 | 84.9 | $0.1 in / $0.3 out |
121
Nemotron 3.5 Super VL (120B A12B)
NVIDIA
46.9
$0.2 in / $1 out
122
Qwen3-235B-A22B-Instruct-2507
Alibaba Cloud / Qwen Team
46.9
$0.09 in / $0.55 out
123
Llama Guard 4 12B
Meta
46.9
$0.18 in / $0.18 out
124