Models.
CSuite supports models across every modality: text, image, audio and video. Run them locally on your machine or in the cloud with your own API keys, private by design. Below are the models bundled with the app out of the box, and you can always add your favorite model inside the app.
Google and Google DeepMind build the Gemini family of multimodal models, the Imagen and Nano Banana image models, the Lyria music models, and the Veo video models.
OpenAI builds GPT, DALL·E, the Sora family, and the open-weight gpt-oss models, and has been a central force behind the modern wave of generative AI.
Alibaba's Tongyi research group publishes the Wan video models and the Qwen family of language models.
ByteDance's AI labs ship the Seedream image models and the Seedance video models, among other Doubao-branded research.
ElevenLabs builds best-in-class voice AI — realistic text-to-speech, voice cloning, music, and sound effects.
Microsoft builds the open-weight Phi family of small language models for reasoning, math and code. Microsoft AI also makes MAI-Transcribe, a speech-to-text model that covers 60 languages.
Meta's FAIR and GenAI labs build the open-weight Llama family of language models — among the most widely used open models for local inference.
MiniMax is a Chinese AI lab whose speech and music models are widely used for multilingual voiceover and generative music.
Liquid develops Liquid Foundation Models (LFM) — a new architecture optimized for efficient long-context reasoning.
Anthropic builds the Claude family of large language models, focused on safety, helpfulness, and honesty.
Black Forest Labs makes the FLUX family of image models — state-of-the-art open-weight image generation.
Kuaishou builds the Kling family of cinematic video models — high-resolution clips with native audio, lip sync, and multi-shot storytelling.
Mistral is a Paris-based lab that ships fast, efficient open-weight language models with strong reasoning per parameter.
IBM Research builds the open-weight Granite family — efficient foundation models tuned for enterprise instruction-following, RAG, code, and tool use.
DeepSeek is a Chinese AI lab whose open-weight models — from the R1 reasoning distills to the V4 mixture-of-experts family — bring frontier-grade reasoning under permissive (MIT) licensing.
xAI builds the Grok family — frontier reasoning LLMs plus Grok Imagine video and voice models, integrated with real-time knowledge from X.
Ideogram builds image models known for accurate typography and design work, with editing from reference images and masks.
NVIDIA builds the open-weight Nemotron family of language and multimodal models, tuned for enterprise reasoning, agentic workflows, and document intelligence.
Runway builds the Gen family of generative video models, pioneering high-fidelity text-to-video and image-to-video for filmmakers and creators.
Deep Reinforce builds the Ornith family — self-improving open-weight (MIT) models for agentic coding, tuned with reinforcement learning on coding and terminal benchmarks.
Fish Audio builds expressive speech models with inline emotion cues, multi-speaker synthesis, and zero-shot voice cloning across 80+ languages.
Pruna AI builds performance-focused image and video models, including the P-Video line, which generates clips with native audio and offers a draft mode for fast, low-cost iteration.
Recraft builds design-first image models — the V4 family, tuned for brand-consistent typography, layout and colour control, plus vector tiers that generate editable SVG artwork rather than raster output.
Z.ai develops the open GLM family of language models, built for long-horizon agentic coding and end-to-end engineering work over very large contexts.
Nace.AI builds the Drex decision models, which return a calibrated probability for each answer to a typed question about a text or JSON state.
vLLM Semantic Router is the vLLM project's open decision layer for models and compute, and publishes the d3 decision models for routing, triage and scoring.
The Allen Institute for AI (Ai2) builds the OLMo family — fully open language models with open weights, training data, and recipes.
Poolside AI builds foundation models for software engineering — the Laguna family, tuned for agentic coding and long-horizon local development.
Moonshot builds the Kimi family — frontier open-weight models with very long context, native visual understanding, and strong agentic coding.
Lightricks builds the LTX family of video models — fast, high-resolution generation with native synchronized audio, first-to-last-frame control, and the open-weight LTX-Video research line.
HeyGen builds AI video tools and models, including the general-purpose HeyGen Video model for short clips from text, images or reference media.
Cloudflare runs a global developer platform and releases open models, including the Clef decision models fine-tuned from Qwen.
Together AI runs a cloud for open-source models and releases its own open-weight models, including the Tev1 decision models fine-tuned from Qwen3.5.
Upstage builds the Solar family of language models and Document AI products, including the Solar Decide decision model.
Perplexity builds an AI answer engine and publishes its own models, including the Decider decision models.
AssemblyAI builds speech AI models and APIs, including the Universal speech-to-text models.
Prism ML builds Bonsai, open vision models trained for ternary weights that run at about half the size of 4-bit models.
Thinking Machines Lab builds open-weight multimodal models, including Inkling and Inkling Small, which read text, images and audio, and Tinker, a fine-tuning API.
Mirelo focuses on generative audio — sound effects and ambient soundscapes driven by text prompts.
Cohere builds enterprise language models — the Command family and the North agent platform — tuned for retrieval-augmented generation, tool use, and secure private deployment.
ACE Studio builds the open ACE-Step music models — full-track generation with lyric editing, remixing, and voice cloning in 50+ languages.
Supertone is a Seoul voice-AI company whose Supertonic text-to-speech models are exported to ONNX and built to run entirely on the user's own device.
TypeSafe AI makes Jev, a System One decision model that answers typed choice, score and yes/no questions with probabilities instead of text.
Bespoke Labs builds data curation tools and open models, including the Nimble decision model fine-tuned from Qwen3.5 9B.
Convai Innovations builds Laya, an open-weight multilingual decision model that answers typed questions about a text in a single forward pass.
Strands Agents is the open-source agent SDK started at AWS; it publishes Strands Decider, a small open decision model for classification, routing and scoring.
Sperid Labs builds Iris, an open 3B image model that generates directly in pixel space without a VAE.
Krea makes AI image and video tools and its own image models, including Krea 3 for generating and editing images.
Moondream builds small, efficient vision-language models for image questions, captions and reading text in images.
Respan builds Span, open-weight decision models that answer typed questions about text and images.
Reflection AI builds open-weight frontier models, including Beam, a reasoning model for coding and long-running agents.
OpenMOSS builds the open MOSS model family, including MOSS-Image, an open 5B text-to-image model.
Gradium builds voice AI models, including Phonon, a family of small text-to-speech models that run on device.
Text
Image
Audio
Video
Decision
Decision models answer typed questions about a text, such as a choice, a yes or no, or a score, with a probability for each option instead of a written reply. CSuite uses them to read what a prompt asks for, so Auto can pick the right model, and to choose between options in workflow Decide nodes.