AI model library
Most talked about this week
by stories in our feed over 7 days- 1GPT-6.2OpenAI99 this week
- 2GPT-6.2 AstraOpenAI95 this week
- 3Claude Fable 5.2Anthropic92 this week
- 4Gemini 4 ArgonGoogle85 this week
- 5Gemini 4Google DeepMind85 this week
- 6Claude Opus 5.5Anthropic37 this week
- 7GPT-6OpenAI29 this week
- 8GPT-6 AstraOpenAI25 this week
- 9Hy4Tencent Hunyuan25 this week
- 10Claude Fable 5.1Anthropic24 this week
Just released
More →GPT-6.2 Astra is a closed-weight large language model from OpenAI, part of the GPT family, designed for general-purpose language tasks.
Anthropic's Claude Fable 5.2 is a proprietary model for demanding reasoning and long-horizon agentic work, accepting text, image and PDF input with a 1M-token context window.
Claude Mythos 5.2 is a closed-weight large language model from Anthropic, released on 2026-11-09 as part of the claude-mythos family.
Muse Ember is Meta's smallest and most efficient reasoning model, built for high-volume agent work such as tool calls, coding subtasks, routing, extraction, and long-document parsing. It accepts text, image, video, audio, and PDF input with a 1M-token context window.
Seedream 5.5 is an image generation model from ByteDance, released on 2026-11-09. It is closed-weight and has no publicly listed license.
Arete-27B is a model from Nous Research, part of the Qwen family, released on November 7, 2026. Weights are not open.
Inkling 2 Small is a closed-weight language model from Thinking Machines, part of the inkling family and released on 2026-11-06. It is intended for general language and chat tasks.
ERNIE 5.5 Flash Base is a closed-weight large language model from Baidu, released on 2026-11-05 as the base variant of the Flash tier in the ERNIE family.
Latest open weights
More →ERNIE 5.5 Flash Base is a closed-weight large language model from Baidu, released on 2026-11-05 as the base variant of the Flash tier in the ERNIE family.
A 0.8B open-weight vision-language model from Typhoon (SCB 10X), released under Apache-2.0, accepting text and image input for OCR-style image understanding and chat.
Typhoon OCR 2 (2B) is a 2.2B-parameter open-weight vision-language OCR model from SCB 10X's Typhoon team. It accepts text and image input for document/text recognition tasks and is released under Apache-2.0 for lightweight deployment.
Qwen3.8-Flash-Next is a closed-weights large language model from Alibaba's Qwen team, released on 2026-10-27.
An OCR model released by Baidu on 2026-10-27, designed for text recognition tasks.
Kimi K3.1 Pro is a proprietary model from Moonshot AI, not released with open weights. No further capabilities or use cases are documented in the available facts.
A fast variant in the M3.1 series released by MiniMax on October 19, 2026. Further details on capabilities are not available from the collected facts.
DeepSeek's large language model supporting text and image input, excels at reasoning and agentic coding with a 1M token context window.
A 0.8B open-weight vision-language model from Typhoon (SCB 10X), released under Apache-2.0, accepting text and image input for OCR-style image understanding and chat.
Typhoon OCR 2 (2B) is a 2.2B-parameter open-weight vision-language OCR model from SCB 10X's Typhoon team. It accepts text and image input for document/text recognition tasks and is released under Apache-2.0 for lightweight deployment.
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.
AI Singapore's 5.1B parameter multimodal model supporting text, image, and audio input, capable of chat and vision tasks.
A multimodal model from AI Singapore Singapore supporting text and image input with chat and vision capabilities.
Apertus by AI Singapore is an 8B parameter open-source LLM with chat capability, licensed under MIT.
A multimodal model developed by AI Singapore supporting text and image input/output with chat and vision capabilities.
All models
sorted by news over 30 daysAnthropic's Claude Fable 5.2 is a proprietary model for demanding reasoning and long-horizon agentic work, accepting text, image and PDF input with a 1M-token context window.
Google's flagship reasoning model for complex software engineering, deep math and science, and long-horizon autonomous agents. It accepts text, image, video, audio and PDF input with a 1,048,576-token context window.
Google DeepMind's large language model in the Gemini series, focusing on multimodal processing capabilities.
Large language model developed by OpenAI, scheduled for release in September 2026, closed-source model.
GPT-6.2 is a large language model in OpenAI's GPT family. The collected facts provide no further details on its capabilities or pricing.
GPT-6.2 Astra is a closed-weight large language model from OpenAI, part of the GPT family, designed for general-purpose language tasks.
OpenAI's GPT-6 Astra, designed for complex reasoning, coding, and document creation with text, image and PDF input.
Kimi K3.1 Pro is a proprietary model from Moonshot AI, not released with open weights. No further capabilities or use cases are documented in the available facts.
Anthropic's Claude Opus is an AI assistant model released in 2026, excelling at complex tasks.
Kimi K3.1 is a large language model from Moonshot AI, part of its Kimi model family. Closed weights.
GPT-6.1 Terra is a closed-weight model in OpenAI's GPT family, released on 2026-10-26. No further specifications or capabilities were provided in the collected facts.
OpenAI's GPT series large language model, with no open weights information available.
DeepSeek-V4.1-Flash-1017 is a closed-weights large language model from DeepSeek, released on 2026-10-17 as a Flash variant of the V4.1 family.
DeepSeek's large language model supporting text and image input, excels at reasoning and agentic coding with a 1M token context window.
A conversational large language model developed by Alibaba Qwen, suitable for various dialogue scenarios.
Large language model from DeepSeek, set to release on September 27, 2026, not open source.
DeepSeek's V4.1-Flash is a closed-source large language model released on September 18, 2026.
A video generation model by Kwai, capable of producing high-quality video content, released on September 29, 2026.
OpenAI's GPT series large language model, scheduled for release on September 29, 2026, not open source.
A fast variant in the M3.1 series released by MiniMax on October 19, 2026. Further details on capabilities are not available from the collected facts.
DeepSeek V4.1 is a large language model from DeepSeek. Limited information is available; it belongs to the DeepSeek model family and its weights are not open.