| Rank | Model | Provider | Score | Benchmarks | Inference | Agentic | Programming | Value | Price |
|---|---|---|---|---|---|---|---|---|---|
| 161 | GPT-4.1 nano gpt-4.1-nano-2025-04-14 multimodalvisionmulti-input reasoning | OpenAI | 34.2 overall | 12.5 | 93.4 | 0.0 | 0.0 | 82.7 | $0.1 in / $0.4 out |
| 162 | Nemotron 3 Nano (30B A3B) nemotron-3-nano-30b-a3b codeprogrammingtool use | NVIDIA | 34.1 overall | 45.4 | 66.0 | 3.3 | 4.4 | 90.9 | $0.06 in / $0.24 out |
| 163 | Qwen3.6-35B-A3B qwen3.6-35b-a3b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 33.7 overall | 55.3 | 0.0 | 15.5 | 26.0 | 0.0 | N/A |
| 164 | MiniMax M1 80K minimax-m1-80k codeprogrammingtool use | MiniMax | 33.6 overall | 24.2 | 84.0 | 20.9 | 19.0 | 41.8 | $0.55 in / $2.2 out |
| 165 | Ministral 8B Instruct ministral-8b-instruct-2410 textinference | Mistral AI | 33.6 overall | 0.0 | 7.0 | 0.0 | 0.0 | 76.1 | $0.1 in / $0.1 out |
| 166 | o3 o3-2025-04-16 multimodalvisionmulti-input reasoning | OpenAI | 33.2 overall | 46.0 | 38.9 | 19.6 | 30.2 | 27.7 | $2 in / $8 out |
| 167 | DeepSeek-V3 deepseek-v3 codeprogrammingtool use | DeepSeek | 33.2 overall | 27.3 | 58.0 | 0.0 | 10.4 | 60.5 | $0.27 in / $1.1 out |
| 168 | DeepSeek-V3.1 deepseek-v3.1 codeprogrammingtool use | DeepSeek | 32.9 overall | 38.4 | 39.8 | 15.2 | 28.3 | 58.8 | $0.27 in / $1 out |
| 169 | DeepSeek R1 Distill Qwen 32B deepseek-r1-distill-qwen-32b textinference | DeepSeek | 32.7 overall | 26.6 | 16.6 | 0.0 | 0.0 | 75.9 | $0.12 in / $0.18 out |
| 170 | DeepSeek-V2.5 deepseek-v2.5 codeprogrammingtool use | DeepSeek | 32.5 overall | 0.0 | 46.5 | 0.0 | 0.9 | 79.7 | $0.14 in / $0.28 out |
| 171 | DeepSeek R1 Distill Llama 70B deepseek-r1-distill-llama-70b textinference | DeepSeek | 32.2 overall | 28.8 | 16.6 | 0.0 | 0.0 | 66.6 | $0.1 in / $0.4 out |
| 172 | Qwen3.5-4B qwen3.5-4b multimodalvisionmulti-input reasoning | Alibaba Cloud / Qwen Team | 32.1 overall | 32.1 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 173 | Claude 3 Haiku claude-3-haiku-20240307 multimodalvisionmulti-input reasoning | Anthropic | 32.0 overall | 5.8 | 61.8 | 0.0 | 0.0 | 57.9 | |
| 174 | Mistral Large 3 (675B Instruct 2512) mistral-large-latest multimodalvisionmulti-input reasoning | Mistral AI | 31.6 overall | 22.2 | 40.1 | 0.0 | 0.0 | 44.5 | |
| 175 | Phi 4 Reasoning Plus phi-4-reasoning-plus textinference | Microsoft | 31.5 overall | 31.5 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 176 | Hermes 3 70B hermes-3-70b textinference | Nous Research | 30.1 overall | 30.1 | 0.0 | 0.0 | 0.0 | 0.0 | N/A |
| 177 | Grok-2 grok-2 multimodalvisionmulti-input reasoning | xAI | 30.1 overall | 27.1 | 38.3 | 0.0 | 0.0 | 25.4 | $2 in / $10 out |
| 178 | GLM-4.7-Flash glm-4.7-flash codeprogrammingtool use | Zhipu AI | 29.9 overall | 38.2 | 29.7 | 11.4 | 20.7 | 72.1 | $0.07 in / $0.4 out |
| 179 | GPT-4o gpt-4o-2024-05-13 multimodalvisionmulti-input reasoning | OpenAI | 29.9 overall | 22.3 | 45.4 | 0.0 | 0.0 | 26.5 | $2.5 in / $10 out |
| 180 | Llama 3.3 70B Instruct llama-3.3-70b-instruct textinference | Meta | 29.9 overall | 19.6 | 21.4 | 0.0 | 0.0 | 72.2 | $0.2 in / $0.2 out |
161
GPT-4.1 nano
OpenAI
34.2
$0.1 in / $0.4 out
162
Nemotron 3 Nano (30B A3B)
NVIDIA
34.1
$0.06 in / $0.24 out
163
Qwen3.6-35B-A3B
Alibaba Cloud / Qwen Team
33.7
N/A
164