AI Models Catalog

Anthropic's most capable widely released model, built for demanding reasoning and long-horizon agentic work, with adaptive thinking and a 1M token context.

Anthropic's fastest, most cost-efficient Claude model, with near-frontier coding at one-third the cost and more than twice the speed of Claude Sonnet 4.

Anthropic's most capable Opus-tier model — highly autonomous, state-of-the-art on long-horizon agentic work, knowledge work, and memory, with 1M token context.

Anthropic's best mix of speed and intelligence — the most agentic Sonnet yet, reaching near-Opus quality on coding and agentic work at Sonnet cost.

Cohere's first Mixture-of-Experts model, released May 2026. 218B total / 25B active parameters, vision input, reasoning, and agentic tool use under Apache 2.0.

Databricks' open Mixture-of-Experts LLM, released March 2024 with 132B total / 36B active. Fully retired from Databricks hosting by December 19, 2025.

DeepSeek's open-weight MoE flagship: V4-Pro (1.6T total / 49B active) and V4-Flash (284B / 13B), with hybrid sparse attention, a 1M context, and an MIT license.

Baidu's closed-weight flagship, released May 9, 2026. A sparse Mixture-of-Experts model with a 128K context — the first ERNIE whose weights stay unpublished.

Google's Gemini 3.5 generation. Gemini 3.6 Flash, shipped July 21, 2026, is the flagship: 1M token context, a March 2026 cutoff, $1.50/$7.50 per 1M.

Google's open-weight Gemma 4 family — five Apache 2.0 models from 2.3B to a 30.7B dense flagship, spanning 140+ languages and native audio, vision, and text.

Zhipu AI's open-weight flagship: 753B parameters, a sparse GLM-MoE-DSA attention stack with IndexShare, a usable 1M token context window, and an MIT license.

OpenAI's flagship family — Sol, Terra and Luna — generally available since July 9, 2026. All three share a 1,050,000 token context and a February 2026 cutoff.

xAI's flagship model, a mixture-of-experts system trained jointly with Cursor. A 500K token context at $2.00 in / $6.00 out per million tokens.

Thinking Machines Lab's first open-weights model: a 975B-parameter multimodal MoE with 41B active, 1M-token context and controllable thinking effort.

Moonshot AI's flagship, announced July 16, 2026. A 2.8T-parameter MoE with 1M context, native vision, always-on thinking, and weights promised by July 27.

Ant Group's open-weight, non-thinking flagship: ~1T parameters with 63B active, a hybrid MLA and Linear Attention stack, a 1M-token context, and an MIT license.

Meta's last open-weight frontier release: Scout and Maverick, natively multimodal Mixture-of-Experts models with 17B active and up to a 10M token context.

Meituan's open-weight flagship, announced June 30, 2026. A 1.6T-parameter MoE with ~48B active parameters, a 1M-token context window, and an MIT license.

Xiaomi's open-weight flagship, released April 22, 2026. A 1.02-trillion-parameter MoE with 42B active, a 1M-token context, and an MIT license.

MiniMax's multimodal flagship, released June 1, 2026. A 428B-parameter MoE with ~23B active, a 1M-token context, and open weights under a community license.

Mistral AI's multimodal model for agentic and coding work: a dense 128B model with a 256K context window, released April 2026 under a modified MIT license.

Meta's frontier model, released April 8, 2026 by Meta Superintelligence Labs. A natively multimodal reasoning model with tool use and multi-agent orchestration.

NVIDIA's open-weight Nemotron 3 flagship: a 550B / 55B-active hybrid Mamba-Transformer MoE with a 1M-token context, NVFP4 pretraining, and an OpenMDW license.

Alibaba Cloud's flagship Qwen model, announced May 20, 2026. A closed-weight, API-only agent with a 1M token context window and hybrid thinking on by default.

ByteDance's current flagship family (Seed 2.1, Doubao), released June 23, 2026. A general agent with multimodal understanding, coding, and GUI task execution.

StepFun's open-weight vision-language flagship, released May 29, 2026. A 198B MoE activating ~11B per token, served over 400 tokens/second under Apache 2.0.

Tencent's open-weight flagship, released July 6, 2026. A 295B-parameter MoE with 21B active, hybrid fast-and-slow thinking, a 256K context, and Apache 2.0.

Cursor's in-house agentic coding model, released May 18, 2026. Built by post-training Moonshot's Kimi K2.5, delivering frontier-adjacent coding at low cost.

ByteDance's dedicated coding model, released February 14, 2026 with the Seed 2.0 family. Combines programming and reasoning with multimodal perception.

ByteDance Seed's standalone vision-language model: a 532M vision encoder with a 20B active-parameter MoE, top results on 38 of 60 public VLM benchmarks.

MiniMax's flagship video model, released October 28, 2025. Closed weights, text-to-video and image-to-video up to 1080p, no audio — still its newest.

Alibaba's video generation model, released June 23, 2026 by the Alibaba Token Hub team. Generates up to 15 seconds of 1080p video with synchronized audio.

Kuaishou's closed-weight video family, launched February 5, 2026. Four SKUs across video and image, clips of 3–15 seconds with natively generated audio.

ByteDance Seed's video model, launched in China February 12, 2026. A unified multimodal audio-video architecture; global rollout was paused in March 2026.

OpenAI's video generation model with synced audio, released September 2025. The consumer app shut down April 2026; the API runs until September 24, 2026.

ShengShu's flagship Vidu Q3 Pro video model, announced January 30, 2026. Generates up to 16 seconds of 1080p video with native audio in a single pass.

Alibaba's open-weights video model, released July 28, 2025 under Apache 2.0. A 27B MoE diffusion transformer with 14B active, plus a 5B variant for an RTX 4090.

HiDream.ai's 8B pixel-level unified transformer, open-sourced under MIT on May 8, 2026. No VAE, no separate text encoder, and generates up to 2,048x2,048.

Tencent's 80B-parameter MoE image model, released September 28, 2025. Natively multimodal and autoregressive, under a restricted community license.

Alibaba's unified image generation and editing model, launched February 10, 2026. Native 2K output, and the line's first flagship shipped with closed weights.

ByteDance's unified multimodal image model. Seedream 5.0 Lite (Feb 13, 2026) added deep thinking and web search grounding; a Pro tier followed in June 2026.

Stability AI's flagship open-weights image model, released October 2024. Three MMDiT variants, still billed as its most powerful image model yet in mid-2026.

Alibaba Tongyi-MAI's 6B open-source image model. Z-Image-Turbo, released November 26, 2025, generates 8-step images inside 16 GB of VRAM under Apache 2.0.

OpenMOSS's open-source speech synthesis model (Feb 6, 2026). An 8B autoregressive TTS system under Apache 2.0, with zero-shot voice cloning and 20 languages.