300 models ranked for open source development. Scored with bonuses for open source licenses (+10), free access (+5), large context (full codebases), large output (complete components), reasoning, and streaming.
| # | Model | Score |
|---|---|---|
| 1 | Nemotron 3 Super (free)NVIDIA | 84 |
| 2 | MiniMax M2.5 (free)MiniMax | 83 |
| 3 | Nemotron Nano 12B 2 VL (free)NVIDIA | 82 |
| 4 | Qwen3.5-9BAlibaba | 85 |
| 5 | Kimi K2.5Moonshot AI | 85 |
| 6 | Qwen3 VL 8B ThinkingAlibaba | 85 |
| 7 | Qwen3 VL 30B A3B ThinkingAlibaba | 85 |
| 8 | GPT-5.4 ProOpenAI | 94 |
| 9 | GPT-5.4OpenAI | 94 |
| 10 | GPT-5.4 MiniOpenAI | 93 |
| 11 | Step 3.5 Flash (free)StepFun | 78 |
| 12 | MiniMax M2.7MiniMax | 83 |
| 13 | GPT-5.2 ProOpenAI | 93 |
| 14 | GPT-5.2OpenAI | 93 |
| 15 | MiMo-V2-FlashXiaomi | 83 |
| 16 | Trinity Miniarcee-ai | 82 |
| 17 | Claude Opus 4.6Anthropic | 92 |
| 18 | GPT-5 ProOpenAI | 92 |
| 19 | Tongyi DeepResearch 30B A3BAlibaba | 82 |
| 20 | Qwen3.5 397B A17BAlibaba | 82 |
| 21 | gpt-oss-safeguard-20bOpenAI | 82 |
| 22 | o3 Deep ResearchOpenAI | 92 |
| 23 | Qwen3 30B A3B Thinking 2507Alibaba | 81 |
| 24 | Claude Opus 4.5Anthropic | 90 |
| 25 | GPT-5OpenAI | 90 |
| 26 | Qwen3.5-122B-A10BAlibaba | 80 |
| 27 | Gemini 3 Flash PreviewGoogle | 89 |
| 28 | Claude Sonnet 4.6Anthropic | 89 |
| 29 | Qwen3.5-27BAlibaba | 79 |
| 30 | Claude Sonnet 4.5Anthropic | 89 |
Open source models come with publicly available code and weights. You can inspect the model architecture, training approach, and data usage. This transparency builds trust and enables community-driven improvements and customizations.
Contribute directly to open source AI projects. Fix bugs, improve performance, add features, and participate in discussions. Open source models benefit from collective intelligence and diverse perspectives across the developer community.
Deploy open source models on your own infrastructure. No vendor lock-in, no monthly subscriptions, complete data privacy. Run models locally on your laptop, server, or cloud infrastructure with full control over updates and deployments.
Fine-tune open source models on your domain-specific data. Adapt models for specialized tasks, proprietary knowledge, or unique vocabularies. Open source enables experimentation without restrictions or vendor approval.
Based on our composite scoring updated hourly, the top-ranked models are shown at the top of this page. Rankings consider benchmarks, pricing, capabilities, and community adoption.
Yes, several models listed on this page offer free tiers or are fully open-source. Look for models marked as Free in the pricing column above.
We use a composite scoring system combining benchmark performance, capability matching, pricing, context window size, and community adoption. Scores are updated hourly.
Rankings refresh every hour using real-time data from benchmarks, API testing, and community metrics. The data shown always reflects the most current performance.