AI model library
IQuest-Q1 is a model from the iquest family, released on September 29, 2026. Its specific用途 is unknown.
Ember-1 is a closed-source large language model developed by Fireworks.
Mk1.5 is an AI model developed by Perceptron, scheduled for release in September 2026.
Agora-2 is an AI model developed by Odyssey, set to be released on September 24, 2026.
An image generation model by inclusionAI designed for design scenarios, not open source.
NVIDIA's audio diarization model for distinguishing different speakers.
UpstageAI's Solar Mini 4 is a compact language model designed for specific task deployment.
Alibaba Qwen3.8-LiveTranslate model, specialized for real-time translation tasks.
Figure's AI model designed for humanoid robots, released on September 18, 2026.
DeepSeek-OCR is an OCR model by DeepSeek for text recognition, released in September 2026.
H3 Max is an AI model developed by fal, scheduled for release on September 17, 2026.
JinaAI's OCR model for text recognition, not open source.
LimiX-2 is an AI model developed by Stable AI and Tsinghua University, released in September 2026.
Cohere's open-weight translation model supporting 50+ languages and locale variants with 32768 token context window.
Nemotron-SEA-LION-v4.8-120B-A12B is an open-source LLM by AI Singapore with chat capabilities.
Nemotron-SEA-LION-v4.8-30B-A3B is an open-weight (MIT) text LLM from AI Singapore, released September 2026 with about 32.9B parameters, suited for chat and general text tasks.
Google's Gemini 3.7 Flash is a high-efficiency model for agentic workflows and multimodal reasoning.
Qwen3.7 Flash is a lightweight multimodal model from Alibaba Qwen handling text, image and video inputs with a 1M-token context window, designed for high-throughput tasks; proprietary and API-only.
xAI's Grok 4.5 supports text, image and PDF input with chat, reasoning, tool use and vision capabilities, featuring a 500k token context window.