--- source: 'https://howaiworks.ai/models' section: models count: 47 --- # Models > All 47 pages in the Models section. Each link is the Markdown twin; drop the `.md` for the HTML page. - [Claude Fable 5.1](https://howaiworks.ai/models/claude-fable.md) — updated 2026-09-06 Anthropic's most capable widely released model: same $10/$50 pricing as Fable 5, cache reads cut 75%, 1M context and always-on adaptive thinking. - [Claude Haiku 4.5](https://howaiworks.ai/models/claude-haiku.md) — updated 2026-09-06 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. - [Claude Opus 5](https://howaiworks.ai/models/claude-opus.md) — updated 2026-07-24 Anthropic's Opus-tier flagship — near-Fable capability at half the price, with adaptive thinking on by default and a 1M token context window. - [Claude Sonnet 5](https://howaiworks.ai/models/claude-sonnet.md) — updated 2026-09-06 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. - [Command A+](https://howaiworks.ai/models/command.md) — updated 2026-07-08 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. - [Composer 2.5](https://howaiworks.ai/models/composer.md) — updated 2026-09-06 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. - [DBRX (Discontinued)](https://howaiworks.ai/models/dbrx.md) — updated 2026-07-08 Databricks' open Mixture-of-Experts LLM, released March 2024 with 132B total / 36B active. Fully retired from Databricks hosting by December 19, 2025. - [DeepSeek V4](https://howaiworks.ai/models/deepseek.md) — updated 2026-09-06 DeepSeek's open-weight V4 family: V4-Pro (1.6T/49B), V4-Flash (284B/13B) and the experimental V4-Flash-Vision-Exp. MIT weights, 1M context. - [Seed 2.1 (Doubao)](https://howaiworks.ai/models/doubao-pro.md) — updated 2026-09-06 ByteDance's flagship Seed 2.1: Pro for deep-thinking agent work, Turbo for scale, plus the weekly-updated Doubao-Seed-Evolving. Closed weights. - [Doubao Seed 2.0 Code](https://howaiworks.ai/models/doubao-seed-code.md) — updated 2026-07-08 ByteDance's dedicated coding model, released February 14, 2026 with the Seed 2.0 family. Combines programming and reasoning with multimodal perception. - [Seed1.5-VL](https://howaiworks.ai/models/doubao-vision.md) — updated 2026-07-08 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. - [ERNIE 5.1](https://howaiworks.ai/models/ernie.md) — updated 2026-09-06 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. - [Gemini 3.8 Flash](https://howaiworks.ai/models/gemini.md) — updated 2026-09-06 Google's Gemini 3.8 Flash, GA September 2, 2026. 1M-token context, $0.75/$3.75 per 1M introductory, plus a gated Flash Cyber variant. - [Gemma 4](https://howaiworks.ai/models/gemma.md) — updated 2026-09-06 Google's open-weight Gemma 4 family — Apache 2.0 models from a 2.3B-effective edge model to a 30.7B dense flagship, with QAT builds for local deployment. - [GLM-5.3](https://howaiworks.ai/models/glm.md) — updated 2026-08-27 Zhipu AI's current line: GLM-5.3, a post-trained coding and security flagship, and GLM-5.3-Flash, a 320B/18B multimodal MoE with hybrid attention. - [GPT-6 Astra](https://howaiworks.ai/models/gpt.md) — updated 2026-09-06 OpenAI's flagship since September 4, 2026. A 1,050,000-token reasoning model for long-horizon agentic work, with native computer use. - [Grok 4.6](https://howaiworks.ai/models/grok.md) — updated 2026-09-06 xAI's agent-focused flagship. 500K context at $2.00 in / $6.00 out per 1M — but a prompt of 200K tokens doubles the rate on the entire request. - [MiniMax H3 (Hailuo 3.0)](https://howaiworks.ai/models/hailuo.md) — updated 2026-09-06 MiniMax's omni-modal video model, announced July 31, 2026: 33B open weights, 2K video up to 15 seconds, with stereo audio generated jointly. - [HappyHorse 1.1](https://howaiworks.ai/models/happyhorse.md) — updated 2026-07-08 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. - [HiDream-O1-Image](https://howaiworks.ai/models/hidream.md) — updated 2026-07-08 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 Hy4 preview](https://howaiworks.ai/models/hunyuan.md) — updated 2026-09-06 Tencent's open-weight flagship, released August 28, 2026. A 770B MoE with 49B active, a 1M-token context, Apache 2.0 — and explicitly a preview. - [HunyuanImage 3.0](https://howaiworks.ai/models/hunyuan-image.md) — updated 2026-07-08 Tencent's 80B-parameter MoE image model, released September 28, 2025. Natively multimodal and autoregressive, under a restricted community license. - [Inkling](https://howaiworks.ai/models/inkling.md) — updated 2026-07-18 Thinking Machines Lab's first open-weights model: a 975B-parameter multimodal MoE with 41B active, 1M-token context and controllable thinking effort. - [Kimi K3](https://howaiworks.ai/models/kimi.md) — updated 2026-09-06 Moonshot AI's flagship: a 2.8T-parameter MoE with 104B active, 1M context and native vision. Open weights shipped July 27, 2026. - [Kling 3.0](https://howaiworks.ai/models/kling.md) — updated 2026-07-08 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. - [Ling-2.6-1T](https://howaiworks.ai/models/ling.md) — updated 2026-07-08 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. - [Llama 4](https://howaiworks.ai/models/llama.md) — updated 2026-09-06 The last release in Meta's Llama line: Scout and Maverick, natively multimodal Mixture-of-Experts models with 17B active and up to a 10M token context. - [LongCat-2.0](https://howaiworks.ai/models/longcat.md) — updated 2026-07-08 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. - [MiMo-V2.5-Pro](https://howaiworks.ai/models/mimo.md) — updated 2026-07-08 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-M3](https://howaiworks.ai/models/minimax-m3.md) — updated 2026-09-06 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 Medium 3.5](https://howaiworks.ai/models/mistral-medium.md) — updated 2026-07-08 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. - [MOSS-TTS](https://howaiworks.ai/models/moss-tts.md) — updated 2026-07-08 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. - [Muse Glimmer](https://howaiworks.ai/models/muse-glimmer.md) — updated 2026-09-06 Meta's Apache 2.0 open-weight 30B multimodal agent model, distilled from Muse Spark and built to run offline on a single 24 GB consumer GPU. - [Muse Spark 1.3](https://howaiworks.ai/models/muse-spark.md) — updated 2026-09-06 Meta's frontier reasoning model, shipped September 2, 2026: a 1M-token multimodal agent model at $1.25/$4.25 per million tokens. - [Nano Banana](https://howaiworks.ai/models/nano-banana.md) — updated 2026-08-02 Google's Gemini image family — Pro, 2, 2 Lite and the legacy original. Prices run $0.0336 to $0.24 per image, and the cheapest one outranks the flagship. - [Nemotron 3 Ultra](https://howaiworks.ai/models/nemotron.md) — updated 2026-09-06 NVIDIA's open Nemotron line: the 550B/55B-active Nemotron 3 Ultra flagship and Nemotron 3.5 Lightning, a 30B/3B-active agent-execution model, both OpenMDW. - [Qwen3.8-Max](https://howaiworks.ai/models/qwen.md) — updated 2026-09-06 Alibaba's flagship Qwen model, released August 3, 2026. A 2.4T-parameter MoE with 95B active, 1M token context, text/image/video input — and open weights. - [Qwen-Image 3.0](https://howaiworks.ai/models/qwen-image.md) — updated 2026-09-06 Alibaba's third-generation image model, announced July 21, 2026: 4,500-token prompts, 10px text, 12 languages — and a second flagship with no weights. - [Seedance 2.5](https://howaiworks.ai/models/seedance.md) — updated 2026-08-02 ByteDance Seed video model, launched July 31, 2026: 30-second audio-video in one pass and 50 reference inputs. No technical report or rate card yet. - [Seedream 5.0](https://howaiworks.ai/models/seedream.md) — updated 2026-07-08 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. - [Sora 2 (Discontinued)](https://howaiworks.ai/models/sora.md) — updated 2026-09-06 OpenAI is retiring Sora: the apps closed April 26, 2026 and the API shuts down September 24, 2026, with no OpenAI successor. - [Stable Diffusion 3.5](https://howaiworks.ai/models/stable-diffusion.md) — updated 2026-09-06 Stability AI's flagship open-weights image model, released October 2024. Three MMDiT variants, still billed as its most powerful image model in September 2026. - [Step-3.7-Flash](https://howaiworks.ai/models/step.md) — updated 2026-07-08 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. - [Veo 3.1](https://howaiworks.ai/models/veo.md) — updated 2026-08-02 Google DeepMind's video model with native audio. Three tiers from $0.05 to $0.60 per second, 4–8 second clips up to 4K, plus video extension. - [Vidu Q3 Pro](https://howaiworks.ai/models/vidu.md) — updated 2026-07-08 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. - [Wan 3.0](https://howaiworks.ai/models/wan.md) — updated 2026-09-06 Alibaba's video family. Wan 3.0 (August 2026) generates 30-second 1080p clips with audio, API-only. Wan 2.2 is still the last Apache 2.0 release. - [Z-Image](https://howaiworks.ai/models/z-image.md) — updated 2026-07-08 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.