AI Models Directory
236 AI models ranked by composite score — LLMs, vision models, audio systems, and multimodal architectures. Compare benchmark performance, pricing, and integration options across providers.
236 models
GPT-5
by OpenAI
OpenAI's frontier model with advanced reasoning, native multimodal understanding, and robust function calling. Designed for complex enterprise workflows and agentic applications.
GPT-4o
by OpenAI
OpenAI's natively multimodal flagship model processing text, image, and audio inputs with a single unified architecture. Delivers GPT-4 Turbo-level intelligence at 2x speed and 50% lower cost, with breakthrough real-time voice capabilities.
Claude 4
by Anthropic
Anthropic's most capable model featuring advanced reasoning, coding, and multimodal capabilities. Excels at complex analysis, agentic tasks, and extended thinking with industry-leading safety.
GPT-4
by OpenAI
OpenAI's breakthrough large language model that demonstrated a significant leap in reasoning and factual accuracy over GPT-3.5. Widely adopted across enterprise and developer workflows for code generation, analysis, and complex problem-solving.
Claude 3.5 Sonnet
by Anthropic
Anthropic's breakout model that surpassed Claude 3 Opus at Sonnet-tier pricing, setting new industry benchmarks for coding. Introduced computer use capability and became the most popular model on the API due to its exceptional intelligence-to-cost ratio.
Midjourney V6
by Midjourney
Midjourney V6 represents a major leap in photorealism, prompt adherence, and artistic coherence, setting a new industry benchmark for AI image generation quality. It introduced native text rendering within images and dramatically improved its understanding of complex, multi-subject prompts.
Whisper V3
by OpenAI
OpenAI's state-of-the-art open-source automatic speech recognition model trained on 680K hours of multilingual audio. Supports 99 languages with near-human accuracy and includes translation, timestamp, and language detection capabilities.
BERT
by Google
BERT (Bidirectional Encoder Representations from Transformers) is Google's landmark 2018 language model that introduced the bidirectional pre-training paradigm using masked language modeling and next sentence prediction. It revolutionized NLP by demonstrating that a single pre-trained model could achieve state-of-the-art results across dozens of downstream tasks with minimal fine-tuning.
Gemini 2.5 Pro
by Google DeepMind
Google DeepMind's flagship thinking model with native multimodal understanding across text, images, audio, and video. Excels at complex reasoning, code generation, and agentic tasks with a million-token context window.
Stable Diffusion XL
by Stability AI
Stability AI's high-resolution image generation model producing photorealistic and artistic images at 1024x1024 resolution. Features a two-stage architecture with a base model and refiner for enhanced detail and compositional quality.
GPT-4 Turbo
by OpenAI
An optimized variant of GPT-4 offering a 128K context window, faster inference, and significantly reduced costs. Introduced JSON mode and improved function calling, making it the preferred GPT-4 variant for production applications.
Llama 3.1 70B
by Meta
Meta's workhorse open-source model with 70B parameters, 128K context window, and native tool-use support. Widely deployed as a cost-effective alternative to proprietary frontier models.
text-embedding-3-large
by OpenAI
OpenAI's most capable text embedding model producing 3072-dimensional vectors with support for Matryoshka representation learning. Offers superior retrieval accuracy over ada-002 with flexible dimensionality reduction for cost-performance trade-offs.
DeepSeek-V3
by DeepSeek
DeepSeek's frontier-class MoE model with 671B total parameters and 37B active, trained using FP8 mixed precision for unprecedented cost efficiency. Matches or exceeds GPT-4o and Claude 3.5 Sonnet on key benchmarks.
o1
by OpenAI
OpenAI's first reasoning model that uses extended internal chain-of-thought before responding. Achieves expert-level performance on competitive math (AIME), PhD-level science (GPQA), and complex coding tasks through deliberative alignment.
ElevenLabs Turbo v2.5
by ElevenLabs
ElevenLabs Turbo v2.5 is a low-latency multilingual text-to-speech model optimized for real-time conversational AI applications, offering sub-400ms first-audio latency while maintaining the high voice cloning fidelity ElevenLabs is known for across 32 languages. It powers a wide range of AI assistant, customer service, and interactive voice applications where natural-sounding, real-time speech is critical.
Llama 3.1 405B
by Meta
The largest openly available language model at 405 billion parameters, rivaling proprietary frontier models in reasoning and knowledge. A landmark release demonstrating open-source models can match closed alternatives.
DALL-E 3
by OpenAI
OpenAI's most advanced image generation model with native ChatGPT integration. Features dramatically improved prompt following, text rendering, and safety mitigations compared to DALL-E 2, generating high-fidelity images from natural language descriptions.
Claude 4 Sonnet
by Anthropic
Anthropic's balanced Claude 4 generation model delivering strong coding and reasoning at competitive pricing. Features improved agentic capabilities and extended thinking, offering a compelling mid-tier option between Haiku and Opus.
Llama 3 70B
by Meta
Meta's high-performance 70B parameter model closing the gap with proprietary frontier models. Achieved competitive results on major benchmarks while remaining fully open-source.
DeepSeek-R1
by DeepSeek
DeepSeek's reasoning-specialized model trained with reinforcement learning to produce detailed chain-of-thought before answering. Achieves performance competitive with OpenAI o1 on math and science benchmarks.
Claude 4.5 Sonnet
by Anthropic
Anthropic's most advanced Sonnet-tier model, combining frontier intelligence with practical speed and cost. Features state-of-the-art coding performance, improved extended thinking, and robust agentic capabilities for complex multi-step workflows.
o3
by OpenAI
OpenAI's most powerful reasoning model and successor to o1, achieving state-of-the-art results on ARC-AGI, GPQA, and competitive math benchmarks. Features adaptive compute that allows users to trade speed for accuracy on the hardest problems.
GPT-2
by OpenAI
GPT-2 is OpenAI's 2019 autoregressive language model that demonstrated for the first time that large-scale unsupervised pre-training on internet text could produce coherent, fluent long-form text generation with zero-shot task performance. Its initial withheld release sparked global debate about AI safety and responsible disclosure of capable AI systems.
Run any model via the AaaS Proxy API
Execute models from HuggingFace, Replicate, fal.ai, Together AI, Groq, and Fireworks through a single unified endpoint.