Track 685 AI models from 226 providers: hosted API models with per-token pricing and context windows, plus open-weight models from Hugging Face with download counts and licenses.
Ordered by current popularity signals, such as OpenRouter weekly usage and Hugging Face downloads.
DeepSeek · Oct 17, 2026
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model...
Z.ai · Aug 26, 2026
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate…
DeepSeek · Sep 10, 2026
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B…
Tencent · Aug 28, 2026
Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks…
OpenAI · Sep 22, 2026
GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and…
Anthropic · Sep 22, 2026
Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large…
MiniMax · Oct 19, 2026
MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and…
stepfun · Oct 8, 2026
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and…
OpenAI · Jul 9, 2026
GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows,…
OpenAI · Sep 29, 2026
GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional...
DeepSeek · Jul 31, 2026
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent…
Z.ai · Aug 18, 2026
GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and…
OpenAI is the AI research company behind ChatGPT and the GPT family of models.
18Anthropic modelsAnthropic is an AI safety company that builds the Claude family of models.
22Google modelsGoogle develops the Gemini models and the open-weight Gemma models through Google DeepMind.
15Meta modelsMeta develops the Llama family of open-weight models.
9Mistral AI modelsMistral AI is a Paris-based company that releases both open-weight and commercial models.
23DeepSeek modelsDeepSeek is a Chinese AI lab known for its open-weight DeepSeek V and R model series.
44Qwen modelsQwen is Alibaba Cloud's family of open-weight and hosted language and multimodal models.
35NVIDIA modelsNVIDIA publishes the Nemotron family and other open models alongside its AI computing platform.
7Microsoft modelsMicrosoft publishes research models and small language models such as the Phi family.
5xAI modelsxAI is the company behind the Grok models.
8Moonshot AI modelsMoonshot AI is the Beijing-based developer of the Kimi models.
17Z.ai modelsZ.ai, formerly Zhipu AI, develops the GLM family of models.
9MiniMax modelsMiniMax is a Shanghai-based AI company building text, speech, and video models.
8Tencent modelsTencent publishes the Hunyuan model family and other open models.
16Liquid AI modelsLiquid AI builds efficient Liquid Foundation Models (LFM) for on-device and edge use.
18inclusionAI modelsinclusionAI is Ant Group's open AI research initiative behind the Ling and Ring models.
9Xiaomi modelsXiaomi develops the MiMo family of models.
8OpenBMB modelsOpenBMB is the open-source community behind the MiniCPM models.
4LG AI Research modelsLG AI Research develops the EXAONE family of models.
Prices are listed per 1 million input tokens in USD as published on OpenRouter. Downloads are Hugging Face downloads over the last 30 days.
Model labs, hubs, and local runtimes for accessing and running AI models.
Listed Aug 2, 2026
Desktop application for discovering, downloading, and running language models locally.
Listed Aug 2, 2026
Local runtime for downloading, managing, and running language models.
Listed Aug 2, 2026
Cloud platform and API for running, fine-tuning, and deploying machine learning models.
Listed Aug 2, 2026
Open collaboration platform and hub for models, datasets, and AI applications.
Listed Aug 2, 2026
Generative AI platform focused on image, video, audio, and 3D models.
Listed Aug 2, 2026
Enterprise AI platform for language models, embeddings, retrieval, and agent workflows.
Listed Aug 2, 2026
Provider of open and commercial language models, assistants, and developer APIs.
Listed Aug 2, 2026
Browser-based workspace for prototyping with Gemini models and the Gemini API.
Listed Aug 2, 2026
AI company and platform behind the Claude family of language models and APIs.
Listed Aug 2, 2026
AI platform providing foundation models, developer APIs, and multimodal capabilities.
The context window is the maximum amount of text, measured in tokens, that a model can consider at once, including your prompt, attached documents, and its own reply. A token is roughly three quarters of an English word.
Hosted models are usually billed per million tokens, with separate prices for input tokens (what you send) and output tokens (what the model writes). Output tokens typically cost more.
API models run on the provider's servers and are billed per use. Open-weight models publish their weights, so you can download and run them on your own hardware, subject to their license.
Models are ordered by public usage signals such as OpenRouter weekly usage rankings and Hugging Face download counts. Models without public evidence stay unscored.