open weights

liquid ai's 587m decision model for text, images and speech. send a state with images or an audio clip and typed questions (choice, score, yes/no); get a probability for every answer. research release

liquid ai's open 3b decision model for text and images. send a state and up to 8 images with typed questions (choice, score, yes/no); get a probability for every answer in one pass, no generation.

perplexity's open decision model for text and images. send a state, up to 4 images and typed questions (choice, score, yes/no); get a calibrated probability for every answer, with no text generation.

the largest decision 2.0 model from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one pass. 29.37b parameters, 32k-token input.

autotrust's open decision model with vision. send text and up to 8 images with typed questions (choice, score, yes/no); get a calibrated probability for every answer, with no text generation.

the 9b decision 2.0 model from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one forward pass, with no text generation. 7.94b parameters, 16,384-token input, about 18.4 ms per question.

the 4b decision 2.0 model from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one forward pass, with no text generation. 4.21b parameters, 16,384-token input, about 12.9 ms per question.

the reasoning variant of decision 2.0 sol 2b, stronger on multi-step problems (arithmetic, code execution, causal and logical questions) at the same speed from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one forward pass, with no text generation. 1.88b parameters, 16,384-token input, about 7.5 ms per question.

the 2b decision 2.0 model from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one forward pass, with no text generation. 1.88b parameters, 16,384-token input, about 7.2 ms per question.

a small, fast decision 2.0 model from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one forward pass, with no text generation. 0.75b parameters, 16,384-token input, about 6.0 ms per question.

the smallest and fastest decision 2.0 model from vllm semantic router. send a state and typed questions (choice, score, yes/no); get a probability for every answer in one forward pass, with no text generation. 0.60b parameters, 8,192-token input, about 4.9 ms per question.

kimi k3 — moonshot's 2.8t open-weight multimodal agentic model with native vision and 1m context for long-horizon coding and reasoning. via the melious api.

glm 5.2 — zai's 744b moe with 1m context and hybrid reasoning for coding and agentic tasks. mit license. via the melious api.

glm 5.2 - zhipu's latest flagship language model with 1m context via openrouter

kimi k2 is the latest, most capable version of open-source model from moonshot ai. built as a thinking agent, scaling multi-step reasoning depth and maintaining stable tool-use across 200–300 sequential calls.

a powerful open-source thinking agent that excels at complex, multi-step problem-solving and consistently uses tools effectively over extended operations.

a powerful language application that excels at multilingual communication and complex task execution, designed for fast performance.

follows precise instructions, excels at function/tool calling, and can process both text and images for tasks like document understanding and content generation.

a fast and versatile tool that can analyze and respond to information from text, images, and audio, designed to run efficiently on small or limited devices.

a high-performance system for complex reasoning, coding, and math problems, featuring strong multilingual support.

a powerful, open-source language system excelling in advanced coding, complex reasoning, and integrating tools for sophisticated tasks.

the flux.2 [klein] is the fastest image models of the flux.2 family. it unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in as low as under a second. built for applications that require real-time image generation without sacrificing quality.

increases video smoothness by using ai to generate and insert extra frames, effectively raising the video's frame rate.

generates audio from any input using a unified framework.