OpenAI
83 active modelsMaker of the GPT and o-series models, for text, reasoning, code and images.
The companies and labs whose models you can use with one FluxCDN API key: 581 active models from 80 makers.
Maker of the GPT and o-series models, for text, reasoning, code and images.
Alibaba's Qwen family: open and hosted models for text, code and reasoning.
Gemini, the open Gemma models, Nano Banana for images and Veo for video.
Paris lab behind Mistral, Devstral and Ministral, open and commercial.
Maker of Claude, a model family for conversation, analysis and code.
Formerly Zhipu AI: the GLM models for chat, code and agents.
NVIDIA's open Nemotron models for reasoning, agents and content safety.
Chinese lab known for efficient open-weight models for reasoning and code.
Meta's models: Muse and the open Llama family.
Language models from Recraft.
The Grok models for chat, reasoning and code, and Grok Imagine for video.
Canadian maker of the Command models for business text, search and tools.
Language models from ElevenLabs.
Shanghai maker of the MiniMax language models and Hailuo video models.
Language models from Microsoft AI.
ByteDance's Seed team: the Seed models for chat, reasoning and code.
Beijing lab behind Kimi, known for long context and agentic coding.
Language models from Perplexity.
Tencent's Hunyuan models, including models for translation.
Language models from VoyageAI by MongoDB.
Amazon's Nova models for fast, low-cost text and multimodal work.
The FLUX lab, known for image generation, with its video model.
Language models from Fish Audio.
The Aion models for reasoning and role-play.
ByteDance's UI-TARS, a model that operates graphical interfaces.
Ant Group's open-model team, maker of the Ling models.
Language models from Sentence Transformers.
Language models from Sourceful.
Xiaomi's MiMo models for reasoning and code.
Alibaba's Wan models, which turn text and images into video.
HeyGen, known for AI avatar videos, with its own video model.
Language models from Krea.
Tokyo research lab that takes its methods from nature.
AI research lab founded in 2025, maker of the Inkling models.
Language models from Unbiased.
Language models from AssemblyAI.
Language models from BAAI.
Language models from Deepgram.
Language models from Intfloat.
Liquid AI's LFM models, light enough to run on a device.
Small, fast vision models that read and describe images.
Language models from Nex AGI.
Independent fine-tunes for creative writing and role-play.
Language models from Baidu.
Small models built for one job, such as Schematron for structured data.
Models that apply code edits fast, for coding tools.
Language models from Particle AI.
Language models from Perceptron.
Language models from PrismML.
Reka's multimodal models for text, images and video.
Models for coding tools: fast code edits and code search.
Llama fine-tunes for role-play and creative writing.
Language models from ScaleDown.
Shanghai lab behind the Step models for reasoning and multimodal work.
Language models from Thenlper.
Korean maker of the Solar models for documents and business text.
Independent group behind the Magnum fine-tunes for creative writing.
Language models from Canopy Labs.
Language models from Elorian.
Language models from Fireworks.
Maker of MythoMax, a long-standing model for stories and role-play.
Language models from hexgrad.
IBM's open Granite models for business tasks.
Language models from KT.
Mancer's Weaver, tuned for storytelling and role-play.
Microsoft's Phi family of small language models.
Language models from Motif Technologies.
Nous Research, maker of the open Hermes models.
Language models from RadixArk.
Language models from Recursive.
US lab building open frontier models.
Language models from Ricursive.
Runway's Gen models for making video.
Language models from Sesame.
Language models from Speechify.
Video models from TBC.
Language models from TypeSafe.
Independent fine-tunes for role-play and stories.
Venice's edition of the Dolphin Mistral fine-tune, with fewer refusals.
Writer's Palmyra models for business writing and agents.