LLM models

OpenAI's gpt-oss-20b is a 22B parameter text generation model hosted on Hugging Face in 8-bit and mxfp4 (4-bit) quantizations. It deploys via the lightweight transformers serve CLI or multiple third-party inference providers, suited for evaluation, experimentation, and moderate loads.

Llama-3.1-8B-Instruct is an 8B-parameter open-source text generation model from Meta, hosted on Hugging Face with a December 2023 knowledge cutoff and JSON-based function calling support.

A 1B-parameter instruction-tuned Llama 3.2 model from unsloth, quantized to 4-bit via bitsandbytes and hosted on Hugging Face. It supports text with an optional ipython environment and has a December 2023 knowledge cutoff.

Qwen3-8B is an 8B-parameter open-source text generation model in the Qwen3 family, hosted on Hugging Face and HuggingChat, updated July 26, 2025, and available through multiple inference providers, though its generated content may be inaccurate.

NVIDIA's Nemotron-Cascade-2-30B-A3B is a 32B parameter text generation model on Hugging Face, supporting configurable reasoning with budget control, tool/function calling, and deployment across multiple inference providers with GGUF quantization.

GLM-5.2 is a 753B-parameter Mixture of Experts text generation model from zai-org, available on Hugging Face with multiple inference providers, tool calling, and up to 1M token context windows.

Hy3-preview is a 299B-parameter text generation model by Tencent, released as a preview on Hugging Face via DeepInfra. It supports tool calling, structured output, configurable reasoning effort, and a 262,144-token context window at $0.14/$0.58 per 1M input/output tokens.

Llama 3.2 1B Instruct is Meta's compact 1B-parameter text generation model hosted on Hugging Face, with a December 2023 knowledge cutoff and JSON-based function/tool calling via a Jinja chat template.

NousCoder-14B is a 15B-parameter text generation model by NousResearch, fine-tuned from Qwen3-14B. It is hosted on the Hugging Face Models hub with 4-bit and 8-bit precision variants, an OpenVINO int4 build, and an SFT derivative.

GigaChat3.1 Audio 10B A1.8B is an ai-sage model on Hugging Face supporting text generation and automatic speech recognition, with a chat template that includes tool rendering functions.

Agents-A1 is a 35B parameter text generation model from InternScience, built on a Mixture of Experts architecture and part of the qwen3_5_moe model family. It is hosted on Hugging Face with GGUF quantized variants and multiple inference providers.

VulnLLM-R-7B is an 8B parameter open-source text generation model from Virtue-AI-HUB, based on the Qwen family and hosted on Hugging Face. It is tagged for security use cases and was updated in December 2025.

AFM-4.5B is a 4.5 billion parameter transformer-based text generation language model by Arcee AI, released July 2025. It supports 11 languages, offers open weights on Hugging Face, and is positioned as a thoughtful conversational assistant.

Midm-2.0-Base-Instruct is an AI-based assistant model developed by KT and published on Hugging Face by K-intelligence. It supports tool/function calling via XML-tagged JSON, outputs structured formats like JSON, SQL, and code, and has a December 2024 knowledge cutoff.

Qwen3-14B is a member of the Qwen3 model family, hosted on Hugging Face and available for direct chat via HuggingChat as an open source conversational AI model.

Gemma 2 2b It Abliterated is a 3B parameter text generation model from the Gemma 2 family, hosted on Hugging Face and compatible with Hugging Face Endpoints, updated July 31, 2024.

Llama-3.1-Nemotron-8B-UltraLong-4M-Instruct is an Nvidia-developed instruction-tuned chat model from the Llama 3.1 family with 8B parameters and a 4M token context window, hosted on Hugging Face with function/tool calling support and a December 2023 knowledge cutoff.

XORTRON.CriminalComputing.LARGE.2026.3 is a large model in the XORTRON family, version 2026.3, published by user darkc0de and hosted on the Hugging Face platform.

Qwen3-1.7B is a model in the Qwen3 family, developed by Qwen and hosted on the Hugging Face platform.

Qwen Studio offers comprehensive functionality spanning chatbot, image and video understanding, image generation, document processing, web search integration, tool utilization, and artifacts.

Introducing Claude 3.5 Sonnet—our most intelligent model yet. Sonnet now outperforms competitor models and Claude 3 Opus on key evaluations, at twice the speed.

DeepSeek V3 is listed in LLM models. Its official website has been verified and the profile is queued for deeper research.

Falcon LLM is a generative large language model (LLM) that helps advance applications and use cases to future-proof our world.

Gemini 1.5 Flash is listed in LLM models. Its official website has been verified and the profile is queued for deeper research.