Added Feb 5, 2026
DeepSeek is the architect of the 'DeepSeek movement,' a fundamental shift in AI development that prioritizes extreme efficiency over raw compute
Why: DeepSeek changed the game by proving that 'expensive' doesn't always mean 'better.' We picked it because it's the first model family to offer true frontier-level reasoning (R1), general intelligence (V3), and advanced vision/OCR (VL2) with an open-weight philosophy and an API price point that makes proprietary models look obsolete.
Added Jan 31, 2026
Kimi k1
Why: Kimi k1.5 is the first model to prove that o1-level reasoning is achievable through efficient, open-weight architectures. We selected it because it consistently matches or exceeds Claude 4.5 in technical benchmarks (AIME, MATH-500) while offering a 2M context window and a significantly lower API price point, making frontier intelligence accessible to everyone.
Added Jan 31, 2026
Qwen 2
Why: We added Qwen 2.5-VL to the Open Frontier movement because it is currently the highest-performing open-weight vision model. It proves that open source can lead in multimodal reasoning, especially for tasks requiring high-resolution OCR and long-form video understanding.
Added Jan 31, 2026
Llama 3
Why: We included Llama 3.2 Vision because it is the most widely supported open multimodal model in the world. Its integration into almost every AI tool and framework makes it the 'default' choice for open-weight vision reasoning.
Added Jan 31, 2026
Pixtral Large is Mistral AI's flagship 124B parameter multimodal model, designed to compete directly with GPT-4o and Claude 3
Why: We added Pixtral Large because it represents the peak of European open-weight AI. It is one of the few open models that truly matches the visual reasoning depth of the top proprietary models, making it essential for the Open Frontier movement.
Added Jan 31, 2026
InternVL 2
Why: We included InternVL 2.5 because it is a consistent leaderboard champion. It often outperforms much larger models in visual reasoning and OCR, making it a critical tool for developers who need GPT-4 level vision without the proprietary lock-in.
Added May 19, 2026
Gemini 3
Why: Gemini 3.5 Flash hits a practical sweet spot for developers and creators who need more capability than entry-level models but do not require the full cost of an Ultra model. Its native multimodal design makes it especially useful for mixed-media tasks.
Added May 19, 2026
Gemini Omni is a single Google model announced at I/O 2026 that can generate and reason across text, images, video, and audio from unified prompts
Why: Gemini Omni represents Google's push toward a single model for all media types. For teams building multimodal products, it simplifies architecture by replacing multiple specialized endpoints with one interface.
Added May 31, 2026
DeepSeek V4-Pro is a high-performance language model from DeepSeek
Why: DeepSeek V4-Pro stands out for combining frontier-level performance with transparent, permanent pricing and open weights. It is a practical choice for teams that want to self-host or avoid unpredictable API costs.
Added May 29, 2026
StepFun Step 3
Why: Step 3.7 Flash offers a competitive Chinese-frontier multimodal model with an MoE architecture that balances capability and inference cost. It is a useful option for vision-language applications and for teams exploring alternatives to US models.
New this month
Added Jul 7, 2026
Microsoft announced a family of MAI-branded models at Build 2026, including MAI-Thinking-1 for reasoning, MAI-Image-2
Why: The MAI family gives Microsoft a cohesive, enterprise-ready AI stack. For organizations already using Microsoft services, these models reduce friction by running inside familiar tools rather than requiring separate platforms.
New this month
Added Jul 7, 2026
Claude Fable 5 is Anthropic's Mythos-class model released on June 9, 2026, focused on creative writing, worldbuilding, and narrative depth
Why: Claude Fable 5 is notable as Anthropic's most experimental creative model. Even with limited availability, it represents an interesting direction for AI-assisted fiction and long-form creative work.
Added Feb 6, 2026
OpenAI's most advanced model to date, featuring a 2M context window and specialized training for complex multi-step reasoning
Why: GPT-5.3 Codex is the 'World's Smartest Planner.' While other AI tools are good at chatting, this one is built for solving huge, difficult problems like planning how a whole software system should work. It has a massive memory (2 million words) so it never loses track of the big picture.
Added Feb 5, 2026
Google's most powerful multimodal model, capable of processing hours of video, thousands of lines of code, or massive document sets in a single prompt
Why: Gemini 3 Ultra offers an unbeatable 10M token context window, allowing it to process entire project histories, hours of video, or massive codebases in a single prompt. Its native multimodal intelligence makes it the only model capable of 'seeing' and 'hearing' complex data sets with the same level of depth as it reads text, providing a unique advantage for large-scale data analysis.
Added May 3, 2026
Nemotron 3 Nano Omni is NVIDIA's compact-but-capable multimodal stack for agentic workflows: one family of endpoints that accept text, images, audio, or video (depending on route) and return text answ...
Why: If your product roadmap says 'agents that see and hear the world,' Omni is built for that integration story—fewer moving parts than bolting Whisper + CLIP + LLM together by hand.