Log in Start Free

AI Models

343+ models from 43 providers — all available through one unified gateway.

Start Exploring

Trending models picked for you — click any to learn more.

GPT-5.2

#1 for reasoning

Gemini 2.5 Pro

Best multimodal

Claude Sonnet 4.5

Top for coding

GPT-5

All-rounder

Gemini 3 Pro

Next-gen preview

GPT-5 Mini

Best value

/

343 models

Trending This Week

Speed benchmarks → See live rankings →

GPT-5.2

OpenAI

#1 for reasoning

Gemini 2.5 Pro

Google

Best multimodal

Claude Sonnet 4.5

Anthropic

Top for coding

GPT-5

OpenAI

All-rounder

Gemini 3 Pro

Google

Next-gen preview

GPT-5 Mini

OpenAI

Best value

Sign up free to start using these models — 100 credits, no credit card.

Get Started Free

GPT-5.2

openai/gpt-5.2

OpenAI's latest flagship with superior language understanding and generation.

GPT-5.2 Pro

openai/gpt-5.2-pro

Extended context and enhanced accuracy for professional workloads.

GPT-5.2 Chat

openai/gpt-5.2-chat

Optimized for multi-turn conversational interactions.

GPT-5.2 Codex

openai/gpt-5.2-codex

Top-tier code generation and software engineering assistant.

GPT-5 Chat

openai/gpt-5-chat

Conversational model with strong reasoning and broad knowledge.

GPT-5

openai/gpt-5

Powerful all-rounder excelling at reasoning, long context, and multimodal tasks.

GPT-5 Pro

openai/gpt-5-pro

Professional-grade accuracy for enterprise use cases.

GPT-5 Mini

openai/gpt-5-mini

Balanced cost and performance for everyday tasks.

GPT-5 Nano

openai/gpt-5-nano

Ultra-fast, cost-efficient model for high-volume workloads.

GPT-5 Codex

openai/gpt-5-codex

Code-specialized model for generation, refactoring, and debugging.

GPT-5.1 Instant

openai/gpt-5.1-instant

Low-latency responses for real-time applications.

GPT-5.1 Thinking

openai/gpt-5.1-thinking

Chain-of-thought reasoning for complex problem solving.

GPT-5.1 Codex

openai/gpt-5.1-codex

Versatile code model with broad language support.

GPT-5.1 Codex Mini

openai/gpt-5.1-codex-mini

Compact code assistant for lightweight development tasks.

GPT-5.1 Codex Max

openai/gpt-5.1-codex-max

Maximum context code model for large codebase analysis.

GPT-5.1

openai/gpt-5.1

Balanced next-gen model with strong general capabilities.

GPT-5.1 Chat

openai/gpt-5.1-chat

Conversational variant of GPT-5.1 optimized for dialogue.

GPT-4.1

openai/gpt-4.1

Reliable general-purpose model with strong instruction following.

GPT-4.1 Mini

openai/gpt-4.1-mini

Cost-effective model for straightforward text tasks.

GPT-4.1 Nano

openai/gpt-4.1-nano

Smallest footprint for classification and simple generation.

GPT-4o

openai/gpt-4o

Fast multimodal model optimized for speed and efficiency.

GPT-4o Mini

openai/gpt-4o-mini

Lightweight variant of GPT-4o for budget-conscious use.

GPT-4o Audio Preview

openai/gpt-4o-audio-preview

GPT-4o with native audio input and output capabilities.

GPT-4o Search Preview

openai/gpt-4o-search-preview

GPT-4o with real-time web search grounding.

GPT-4o Mini Search

openai/gpt-4o-mini-search

Compact search-grounded model for quick lookups.

GPT-4 Turbo

openai/gpt-4-turbo

High-capability model with 128K context and vision support.

GPT-4

openai/gpt-4

Original GPT-4 with strong reasoning and broad knowledge.

GPT-3.5 Turbo

openai/gpt-3.5-turbo

Fast and affordable model for simple tasks and high-volume use.

o4-mini

openai/o4-mini

Compact reasoning model with strong analytical capabilities.

o3

openai/o3

Advanced reasoning model for math, science, and logic.

o3-mini

openai/o3-mini

Efficient reasoning at reduced cost.

o3-pro

openai/o3-pro

Professional reasoning with extended thinking time.

o1

openai/o1

Original reasoning model with deliberate chain-of-thought.

Codex Mini

openai/codex-mini

Lightweight coding assistant for quick completions.

GPT-OSS 120B

openai/gpt-oss-120b

Open-source 120B parameter model for research and deployment.

GPT-OSS 20B

openai/gpt-oss-20b

Compact open-source model for self-hosted applications.

DALL-E 2

openai/dall-e-2

Second-gen image generation with editing and variation support.

DALL-E 3

openai/dall-e-3

High-quality image generation with superior prompt adherence.

GPT-Image-1

openai/gpt-image-1

OpenAI's native image generation model integrated with GPT.

GPT-Image-1 Mini

openai/gpt-image-1-mini

Compact image generation for quick creative iterations.

GPT-Image-1.5

openai/gpt-image-1.5

Enhanced image generation with improved quality and coherence.

TTS-1

openai/tts-1

Standard text-to-speech with natural sounding voices.

TTS-1 HD

openai/tts-1-hd

High-definition text-to-speech with superior audio quality.

GPT-4o Mini TTS

openai/gpt-4o-mini-tts

Compact TTS model powered by GPT-4o Mini architecture.

Claude Sonnet 4.5

anthropic/claude-sonnet-4.5

Balanced performance and speed for general-purpose use.

Claude Opus 4.5

anthropic/claude-opus-4.5

High-capability model for deep analysis and creative work.

Claude Opus 4.1

anthropic/claude-opus-4.1

Strong reasoning with extended context support.

Claude Opus 4

anthropic/claude-opus-4

Reliable workhorse for enterprise applications.

Claude Sonnet 4

anthropic/claude-sonnet-4

Fast and capable for everyday analytical tasks.

Claude 3.7 Sonnet

anthropic/claude-3.7-sonnet

Previous-gen Sonnet with proven reliability.

Claude Haiku 4.5

anthropic/claude-haiku-4.5

Ultra-fast responses for high-throughput applications.

Claude 3.5 Haiku

anthropic/claude-3.5-haiku

Quick and cost-efficient for simple tasks.

Claude 3.5 Haiku (Oct 2024)

anthropic/claude-3.5-haiku-20241022

Dated release of Claude 3.5 Haiku for reproducibility.

Claude 3.5 Sonnet

anthropic/claude-3.5-sonnet

Well-rounded model excelling at analysis, writing, and coding.

Claude 3 Haiku

anthropic/claude-3-haiku

Fastest Claude for lightweight interactions.

Claude 3 Haiku (Mar 2024)

anthropic/claude-3-haiku-20240307

Dated release of Claude 3 Haiku for consistent behavior.

Claude 3 Opus

anthropic/claude-3-opus

Legacy Opus with strong reasoning capabilities.

Claude Sonnet 4.5 (Vertex)

anthropic/claude-sonnet-4.5-vertex

Sonnet 4.5 available via Google Vertex AI.

Claude Sonnet 4 (Vertex)

anthropic/claude-sonnet-4-vertex

Claude Sonnet 4 available via Google Vertex AI.

Gemini 3 Flash

google/gemini-3-flash

Google's fastest next-gen model for efficient tasks.

Gemini 3 Pro Preview

google/gemini-3-pro-preview

Next-generation Pro model with advanced capabilities.

Gemini 3 Pro Image

google/gemini-3-pro-image

Image generation from Google's latest Pro model.

Gemini 3 Pro Image Edit

google/gemini-3-pro-image-edit

AI-powered image editing with Gemini 3 Pro.

Gemini 2.5 Flash

google/gemini-2.5-flash

Balanced cost and capability for multimodal reasoning.

Gemini 2.5 Flash Lite

google/gemini-2.5-flash-lite

Cheapest Gemini for classification and summarization.

Gemini 2.5 Flash Lite Preview

google/gemini-2.5-flash-lite-preview

Preview release of Flash Lite for early access testing.

Gemini 2.5 Flash Image

google/gemini-2.5-flash-image

Fast image generation from text prompts.

Gemini 2.5 Flash Image Edit

google/gemini-2.5-flash-image-edit

AI-powered image editing with Gemini 2.5 Flash.

Gemini 2.5 Pro

google/gemini-2.5-pro

Top-tier Gemini for complex reasoning and large context.

Gemini 2.0 Flash

google/gemini-2.0-flash

Previous-gen Flash with proven speed.

Gemini 2.0 Flash Lite

google/gemini-2.0-flash-lite

Lightweight model for simple workloads.

Gemini 2.0 Flash Exp

google/gemini-2.0-flash-exp

Experimental Flash variant with cutting-edge features.

Gemma 3 27B

google/gemma-3-27b

Open-weight 27B model for self-hosted deployments.

Gemma 3 12B

google/gemma-3-12b

Compact open-weight model balancing quality and efficiency.

Gemma 3 4B

google/gemma-3-4b

Ultra-compact model for edge and mobile deployment.

Gemma 3n 4B

google/gemma-3n-4b

Next-gen compact Gemma optimized for on-device use.

Imagen 4.0

google/imagen-4.0

Google's premier image generation model.

Imagen 4.0 Ultra

google/imagen-4.0-ultra

Highest fidelity image generation from Google.

Imagen 4.0 Fast

google/imagen-4.0-fast

Quick image generation for rapid iteration.

Imagen 3.0

google/imagen-3.0

Previous-gen Imagen with proven photorealistic quality.

Veo 3.1

google/veo-3.1-generate-001

Google's latest video generation model with top-tier cinematic quality and audio.

Veo 3.1 Fast

google/veo-3.1-fast-generate-001

Speed-optimized Veo 3.1 for rapid video iteration.

Veo 3.1 I2V

google/veo-3.1-i2v

Image-to-video generation with Veo 3.1 quality.

Veo 3.1 R2V

google/veo-3.1-r2v

Reference-to-video generation for consistent character animation.

Veo 3.1 Extend

google/veo-3.1-extend

Extend existing video clips with coherent continuation.

Veo 3.0

google/veo-3.0-generate-001

High-quality video generation with strong motion coherence and audio support.

Veo 3.0 Fast

google/veo-3.0-fast-generate-001

Fast variant of Veo 3.0 for quick creative exploration.

Veo 3.0 I2V

google/veo-3.0-i2v

Image-to-video generation with Veo 3.0.

Veo 2 T2V

google/veo-2-t2v

Text-to-video generation with Veo 2.

Veo 2 I2V

google/veo-2-i2v

Image-to-video animation with Veo 2.

Lyria 2

google/lyria-2

Google's music generation model for creating instrumentals and compositions.

Gemini 3 Flash Preview

google/gemini-3-flash-preview

Preview release of Gemini 3 Flash for early access.

Gemini 3 Pro Image Preview

google/gemini-3-pro-image-preview

Preview of next-gen Gemini image generation.

Gemini 2.5 Flash Preview

google/gemini-2.5-flash-preview

Preview release of Gemini 2.5 Flash.

Gemini 2.5 Pro Preview

google/gemini-2.5-pro-preview

Preview release of Gemini 2.5 Pro.

Grok-4.1 Fast (Non-Reasoning)

x-ai/grok-4.1-fast-nr

Fastest Grok for direct, non-reasoning responses.

Grok-4.1 Fast (Reasoning)

x-ai/grok-4.1-fast-r

Fast reasoning with chain-of-thought capability.

Grok-4 Fast (Non-Reasoning)

x-ai/grok-4-fast-nr

Quick responses without reasoning overhead.

Grok-4 Fast (Reasoning)

x-ai/grok-4-fast-r

Balanced speed and reasoning depth.

Grok-4

x-ai/grok-4

xAI's flagship model with broad knowledge and wit.

Grok-3

x-ai/grok-3

Powerful conversational AI with real-time data access.

Grok-3 Beta

x-ai/grok-3-beta

Beta variant of Grok-3 with experimental features.

Grok-3 Mini

x-ai/grok-3-mini

Compact Grok for cost-effective deployments.

Grok-3 Mini Fast

x-ai/grok-3-mini-fast

Ultra-fast mini model for real-time use.

Grok-3 Fast

x-ai/grok-3-fast

Speed-optimized Grok-3 variant.

Grok-3 Vision

x-ai/grok-3-vision

Grok-3 with enhanced image understanding.

Grok-2 Vision

x-ai/grok-2-vision

Multimodal Grok with image understanding.

Grok Code Fast 1

x-ai/grok-code-fast-1

Speed-optimized code generation from xAI.

Grok Imagine Image

x-ai/grok-imagine-image

Image generation powered by xAI.

Grok Imagine Image Pro

x-ai/grok-imagine-image-pro

Professional-grade image generation from xAI.

Grok Imagine Video

x-ai/grok-imagine-video

Text-to-video generation powered by xAI.

DeepSeek V3.2

deepseek/deepseek-v3.2

Latest high-performance open-source model.

DeepSeek V3.2 Thinking

deepseek/deepseek-v3.2-thinking

Extended reasoning with chain-of-thought.

DeepSeek V3.2 Speciale

deepseek/deepseek-v3.2-speciale

Specialized variant tuned for specific domains.

DeepSeek V3.2-Exp Thinking

deepseek/deepseek-v3.2-exp-thinking

Experimental thinking mode with deeper reasoning.

DeepSeek V3.2-Exp Non-Thinking

deepseek/deepseek-v3.2-exp-non-thinking

Fast experimental variant without extended reasoning.

DeepSeek V3.1

deepseek/deepseek-v3.1

Reliable open-source model for general tasks.

DeepSeek V3.1 Chat

deepseek/deepseek-v3.1-chat

Chat-optimized variant of V3.1 for conversational use.

DeepSeek V3.1 Reasoner

deepseek/deepseek-v3.1-reasoner

Dedicated reasoning variant with extended analysis.

DeepSeek V3.1 Terminus

deepseek/deepseek-v3.1-terminus

Optimized variant for production workloads.

DeepSeek V3.1 Terminus Reasoning

deepseek/deepseek-v3.1-terminus-reasoning

Terminus with extended chain-of-thought reasoning.

DeepSeek V3.1 Terminus NR

deepseek/deepseek-v3.1-terminus-non-reasoning

Terminus variant without reasoning overhead for speed.

DeepSeek V3

deepseek/deepseek-v3

Strong open-source alternative to proprietary models.

DeepSeek R1

deepseek/deepseek-r1

Research-focused model with advanced chain-of-thought reasoning.

Llama 4 Maverick

meta/llama-4-maverick

Meta's most capable open-weight model.

Llama 4 Scout

meta/llama-4-scout

Efficient Llama 4 variant for fast inference.

Llama 3.3 70B

meta/llama-3.3-70b

Strong 70B model with broad multilingual support.

Llama 3.3 70B Versatile

meta/llama-3.3-70b-versatile

Versatile variant optimized for diverse task types.

Llama 3.2 90B

meta/llama-3.2-90b

Large-scale model for demanding tasks.

Llama 3.2 11B

meta/llama-3.2-11b

Mid-size model balancing performance and cost.

Llama 3.2 3B

meta/llama-3.2-3b

Compact model for edge deployment.

Llama 3.2 3B Instruct Turbo

meta/llama-3.2-3b-instruct-turbo

Speed-optimized instruction-tuned compact model.

Llama 3.2 1B

meta/llama-3.2-1b

Tiny model for on-device applications.

Llama 3.1 405B Instruct Turbo

meta/llama-3.1-405b-instruct-turbo

Massive 405B model with turbo inference speed.

Llama 3.1 70B

meta/llama-3.1-70b

Proven 70B model with strong instruction following.

Llama 3.1 8B

meta/llama-3.1-8b

Lightweight model for simple tasks and fine-tuning.

Llama 3 8B Instruct Lite

meta/llama-3-8b-instruct-lite

Lightweight instruction-tuned model for cost-efficient use.

Devstral 2

mistral/devstral-2

Top-tier agentic coding model with 256K context, multi-file understanding, and autonomous planning.

Devstral Small 2

mistral/devstral-small-2

Second-gen compact code model with improved contextual awareness.

Devstral Small

mistral/devstral-small

Original lightweight code assistant optimized for low-latency autocomplete.

Mistral Large 3

mistral/mistral-large-3

Flagship 128K-context enterprise model with strong multilingual fluency.

Mistral Large 2

mistral/mistral-large-2

Previous-gen flagship with proven enterprise reliability.

Mistral Medium

mistral/medium

Sweet-spot model delivering near-flagship reasoning quality at lower cost.

Mistral Small

mistral/small

Efficiency-first model with sub-100ms latency for high-volume tasks.

Mistral Nemo

mistral/nemo

12B open-weight model built with Nvidia, Apache 2.0 licensed.

Codestral

mistral/codestral

Dedicated 256K-context coding model with fill-in-the-middle completion.

Pixtral Large

mistral/pixtral-large

Flagship multimodal model with high-res vision and OCR-quality extraction.

Pixtral 12B

mistral/pixtral-12b

Compact 12B open-weight multimodal model for visual Q&A.

Magistral Small

mistral/magistral-small

Entry-level reasoning model with visible chain-of-thought traces.

Magistral Medium

mistral/magistral-medium

Advanced reasoning model with deep multi-step analysis.

Ministral 8B

mistral/ministral-8b

Ultra-compact 8B model for edge, mobile, and on-device AI.

Mixtral 8x22B

mistral/mixtral-8x22b

Sparse mixture-of-experts with exceptional throughput and quality.

Mixtral 8x7B Instruct

mistral/mixtral-8x7b-instruct

Instruction-tuned MoE model with strong performance per cost.

Mistral 7B Instruct

mistral/mistral-7b-instruct

Compact 7B instruction-tuned model for self-hosted deployments.

Mistral Tiny

mistral/mistral-tiny

Smallest Mistral model for ultra-low-latency classification tasks.

Mistral OCR

mistral/mistral-ocr

Specialized model for optical character recognition and document extraction.

Mistral Embed

mistral/mistral-embed

High-quality text embeddings from Mistral.

Mistral Large

mistral/mistral-large

Mistral's flagship large model with strong multilingual and reasoning capabilities.

Qwen3 Max (Thinking)

alibaba/qwen3-max-thinking

Alibaba's strongest reasoning model with extended thinking.

Qwen3 Max

alibaba/qwen3-max

Flagship Qwen model for demanding tasks.

Qwen3 Max Preview

alibaba/qwen3-max-preview

Preview release of Qwen3 Max for early testing.

Qwen3 Max Instruct

alibaba/qwen3-max-instruct

Instruction-tuned Qwen3 Max for precise task following.

Qwen3 Next 80B

alibaba/qwen3-next-80b

Next-gen 80B model pushing Alibaba's frontier capabilities.

Qwen3 Coder

alibaba/qwen3-coder

Specialized coding model from Alibaba.

Qwen3 Coder Plus

alibaba/qwen3-coder-plus

Enhanced coding with larger context.

Qwen3 Coder Next

alibaba/qwen3-coder-next

Next-gen coding capabilities.

Qwen3 Omni 30B

alibaba/qwen3-omni-30b

Multimodal omni model handling text, image, audio, and video.

Qwen3 VL 32B Instruct

alibaba/qwen3-vl-32b-instruct

Vision-language model with instruction tuning.

Qwen3 VL Thinking

alibaba/qwen3-vl-thinking

Vision-language model with reasoning capabilities.

Qwen3 VL Instruct

alibaba/qwen3-vl-instruct

Vision-language model for instruction following.

Qwen-3 235B

alibaba/qwen-3-235b

Massive 235B parameter model for complex tasks.

Qwen-3 32B

alibaba/qwen-3-32b

Balanced size and capability.

Qwen-3 30B

alibaba/qwen-3-30b

Efficient model for general use.

Qwen-3 14B

alibaba/qwen-3-14b

Mid-size model for cost-effective deployment.

Qwen 2.5 72B Instruct Turbo

alibaba/qwen-2.5-72b-instruct-turbo

Fast 72B instruction-tuned model for high-quality responses.

Qwen 2.5 7B Instruct Turbo

alibaba/qwen-2.5-7b-instruct-turbo

Compact instruction-tuned model for cost-efficient use.

Qwen3 Plus

alibaba/qwen3-plus

Enhanced Qwen3 for demanding workloads.

Qwen3 Turbo

alibaba/qwen3-turbo

Speed-optimized Qwen3 variant.

Qwen Image

alibaba/qwen-image

Text-to-image generation from Alibaba's Qwen family.

Qwen Image Edit

alibaba/qwen-image-edit

AI-powered image editing with Qwen.

Z-Image Turbo

alibaba/z-image-turbo

Fast image generation optimized for throughput.

Wan 2.2 T2I Plus

alibaba/wan-2.2-t2i-plus

Text-to-image generation with enhanced quality.

Wan 2.2 T2I Flash

alibaba/wan-2.2-t2i-flash

Fast text-to-image for rapid iteration.

Wan 2.5 T2I

alibaba/wan-2.5-t2i

Next-gen text-to-image from the Wan architecture.

Wan 2.6 Image

alibaba/wan-2.6-image

Latest Wan image generation model.

Wan V2.2 T2V

alibaba/wan-v2.2-t2v

Text-to-video generation with Wan V2.2.

Wan V2.5 T2V

alibaba/wan-v2.5-t2v

Improved text-to-video from Wan V2.5.

Wan V2.6 (Text-to-Video)

alibaba/wan-v2.6-t2v

Alibaba's open-source text-to-video model with strong quality.

Wan V2.6 (Image-to-Video)

alibaba/wan-v2.6-i2v

Image-to-video generation from Alibaba's Wan architecture.

Qwen Text Embedding V3

alibaba/qwen-text-embedding-v3

High-quality text embedding from the Qwen family.

Qwen Text Embedding V4

alibaba/qwen-text-embedding-v4

Latest Qwen embedding model with improved retrieval quality.

GTE Large EN V1.5

alibaba/gte-large-en-v1.5

Alibaba's English text embedding model.

GTE Multilingual

alibaba/gte-multilingual

Multilingual embedding from Alibaba.

GLM-5

zai/glm-5

ZAI's most powerful general language model.

GLM-4.7

zai/glm-4.7

Advanced model with strong multilingual support.

GLM-4.7 Flash

zai/glm-4.7-flash

Speed-optimized variant of GLM-4.7.

GLM-4.6

zai/glm-4.6

Reliable model for general-purpose tasks.

GLM-4.6V

zai/glm-4.6v

Vision-enabled variant with image understanding.

GLM-4.5

zai/glm-4.5

Balanced performance and efficiency.

GLM-4.5 Air

zai/glm-4.5-air

Lightweight model for quick responses.

GLM-4.5V

zai/glm-4.5v

Multimodal model with vision capabilities.

GLM-Z1 Flash

zai/glm-z1-flash

Fast reasoning model from ZAI.

GLM-Z1 Air

zai/glm-z1-air

Lightweight GLM reasoning variant.

CogView-4

zai/cogview-4

ZAI's latest image generation model.

Kimi K2.5

moonshotai/kimi-k2.5

MoonshotAI's latest flagship model.

Kimi K2

moonshotai/kimi-k2

Strong general-purpose model with deep reasoning.

Kimi K2 Thinking

moonshotai/kimi-k2-thinking

Extended reasoning variant of Kimi K2.

Kimi K2 Turbo

moonshotai/kimi-k2-turbo

Speed-optimized Kimi for real-time use.

Kimi K2 Thinking Turbo

moonshotai/kimi-k2-thinking-turbo

Fast reasoning with turbo performance.

Kimi K2-0905

moonshotai/kimi-k2-0905

Specialized variant for specific workloads.

MiniMax M2.5

minimax/m2.5

MiniMax's latest and most capable model.

MiniMax M2.1

minimax/m2.1

Reliable model for general tasks.

MiniMax M2.1 Lightning

minimax/m2.1-lightning

Ultra-fast variant for latency-sensitive apps.

MiniMax M2

minimax/m2

Solid foundation model from MiniMax.

MiniMax Text-01

minimax/text-01

First-gen text model with proven reliability.

MiniMax M1

minimax/m1

Original MiniMax model for general use.

MiniMax Music 1.5

minimax/music-1.5

Music generation model producing full tracks.

MiniMax Music 2.0

minimax/music-2.0

Next-gen music generation with enhanced quality.

MiniMax Speech 02 HD

minimax/speech-02-hd

High-definition text-to-speech with natural voices.

MiniMax M2.5 Image

minimax/m2.5-image

Image generation from MiniMax.

Seed 1.8

bytedance/seed-1.8

ByteDance's latest language model.

Seed 1.6

bytedance/seed-1.6

Previous-gen Seed with proven reliability.

Seedream 3.0

bytedance/seedream-3.0

Text-to-image generation with artistic quality.

Seedream 4.0

bytedance/seedream-4.0

Enhanced image generation with improved coherence.

Seedream 4.5

bytedance/seedream-4.5

Latest Seedream with top-tier photorealism.

Seedream USO

bytedance/seedream-uso

Ultra-style-optimized image generation.

Seedance 1.0 Lite T2V

bytedance/seedance-1.0-lite-t2v

Lightweight text-to-video generation.

Seedance 1.0 Pro T2V

bytedance/seedance-1.0-pro-t2v

Professional-grade text-to-video from ByteDance.

Seedance 1.0 Lite I2V

bytedance/seedance-1.0-lite-i2v

Lightweight image-to-video animation.

Seedance 1.0 Pro I2V

bytedance/seedance-1.0-pro-i2v

Pro-tier image-to-video from ByteDance.

OmniHuman

bytedance/omnihuman

Human-centric video generation for avatars and motion.

Nova Micro

amazon/nova-micro

Amazon's compact model for simple tasks.

Nova Lite

amazon/nova-lite

Lightweight model for cost-efficient deployment.

Nova Pro

amazon/nova-pro

Professional-grade model from Amazon.

Nova 2 Lite

amazon/nova-2-lite

Next-gen lightweight Amazon model.

Titan Embed Text V2

amazon/titan-embed-text-v2

Amazon's text embedding model for search and retrieval.

Command-A

cohere/command-a

Enterprise-focused model built for RAG and tool use.

Command R Plus

cohere/command-r-plus

Cohere's most capable model for RAG and enterprise use.

Command R

cohere/command-r

Fast RAG-optimized model from Cohere.

Embed V4.0

cohere/embed-v4.0

Cohere's latest embedding model with state-of-the-art retrieval.

Embed English V3.0

cohere/embed-english-v3.0

English-optimized embedding from Cohere.

Embed Multilingual V3.0

cohere/embed-multilingual-v3.0

Multilingual embedding supporting 100+ languages.

Nemotron Ultra

nvidia/nemotron-ultra

Nvidia's most powerful language model.

Nemotron Mini

nvidia/nemotron-mini

Compact Nvidia model for efficient inference.

Nemotron Nano 9B V2

nvidia/nemotron-nano-9b-v2

Second-gen 9B model for on-device and edge use.

Nemotron Nano 12B V2 VL

nvidia/nemotron-nano-12b-v2-vl

Vision-language variant of Nemotron Nano 12B.

Llama 3.1 Nemotron 70B

nvidia/llama-3.1-nemotron-70b

Nvidia-tuned Llama 3.1 70B with enhanced performance.

MiMo V2 Flash

xiaomi/mimo-v2-flash

Xiaomi's fast and efficient language model.

Longcat 128K

meituan/longcat-128k

Ultra-long context model supporting 128K tokens.

Longcat 32K

meituan/longcat-32k

Extended context model for long documents.

Step-3

meituan/step-3

Meituan's flagship large language model.

V0 1.5 MD

vercel/v0-1.5-md

Vercel's AI for generating UI components and code.

Morph V3

morph/v3

Versatile model for text generation and analysis.

Morph V3 Mini

morph/v3-mini

Compact Morph model for lightweight tasks.

Arcee Trinity

arcee/trinity

Merged model combining multiple expert capabilities.

Arcee Virtuoso Medium V2

arcee/virtuoso-medium-v2

Enterprise-grade merged model from Arcee.

Intellect 3

prime-intellect/intellect-3

Decentralized training model with unique architecture.

Prime Intellect

Mercury

inception/mercury

High-speed inference model with low latency.

Mercury Coder

inception/mercury-coder

Ultra-fast code generation from Inception.

KAT-Coder

kwaipilot/kat-coder

Specialized coding model from KwaiPilot.

Flux Pro 1.1 Ultra

bfl/flux-pro-1.1-ultra

Highest fidelity image generation from Black Forest Labs.

Flux Pro 1.1

bfl/flux-pro-1.1

Professional-grade image generation.

Flux Dev

bfl/flux-dev

Developer-friendly image generation for rapid prototyping.

Flux Schnell

bfl/flux-schnell

Ultra-fast image generation for real-time use.

Flux Realism

bfl/flux-realism

Photorealism-focused image generation model.

Flux SRPO

bfl/flux-srpo

Super-resolution and prompt-optimized generation.

Flux 2 Max

bfl/flux-2-max

Maximum quality image generation.

Flux 2 Pro

bfl/flux-2-pro

Pro-tier image generation with fine control.

Flux 2

bfl/flux-2

Second-gen Flux with improved quality and speed.

Flux 2 Edit

bfl/flux-2-edit

Image editing and inpainting with Flux 2.

Flux 2 LoRA

bfl/flux-2-lora

LoRA fine-tuning support for customized generation.

Flux 2 Flex

bfl/flux-2-flex

Flexible image generation for varied styles.

Flux 2 Klein 9B

bfl/flux-2-klein-9b

Compact 9B image model for fast generation.

Flux 2 Klein 4B

bfl/flux-2-klein-4b

Ultra-compact 4B image model.

Flux Kontext Pro

bfl/flux-kontext-pro

Context-aware image editing and generation.

Flux Kontext Max

bfl/flux-kontext-max

Maximum capability context-aware image generation.

Flux Pro 1.0 Fill

bfl/flux-pro-1.0-fill

Inpainting and outpainting specialist.

Recraft V2

recraft/v2

Design-focused image generation with brand consistency.

Recraft V3

recraft/v3

Latest Recraft model with enhanced style control.

Kling V3.0 (Text-to-Video)

klingai/kling-v3.0-t2v

KlingAI's flagship text-to-video model with cinematic motion.

Kling V3.0 (Image-to-Video)

klingai/kling-v3.0-i2v

Animate still images into fluid video with Kling V3.0.

Kling V2.6 (Text-to-Video)

klingai/kling-v2.6-t2v

Cost-effective text-to-video generation from KlingAI.

Kling V2.6 (Image-to-Video)

klingai/kling-v2.6-i2v

Image-to-video animation with Kling V2.6.

Kling V2.5 T2V

klingai/kling-v2.5-t2v

Text-to-video with Kling V2.5 architecture.

Kling V2.5 I2V

klingai/kling-v2.5-i2v

Image-to-video with Kling V2.5.

Kling V2.1 T2V

klingai/kling-v2.1-t2v

Reliable text-to-video from the V2.1 generation.

Kling V2.0 T2V

klingai/kling-v2.0-t2v

Previous-gen text-to-video with proven quality.

Kling V1.6 T2V

klingai/kling-v1.6-t2v

Legacy text-to-video model at reduced cost.

Kling V1 T2V

klingai/kling-v1-t2v

Original Kling video generation model.

Kling Avatar Standard

klingai/kling-avatar-standard

Avatar generation with standard quality and speed.

Kling Avatar Pro

klingai/kling-avatar-pro

Professional avatar generation with high fidelity.

Kling Image O1

klingai/kling-image-o1

Image generation with reasoning-enhanced prompting.

Kling Video O1

klingai/kling-video-o1

Video generation with reasoning-enhanced quality.

Text Embedding 3 Large

openai/text-embedding-3-large

OpenAI's most capable text embedding model.

Text Embedding 3 Small

openai/text-embedding-3-small

Cost-efficient embedding for retrieval and classification.

Text Embedding Ada 002

openai/text-embedding-ada-002

Legacy embedding model with broad adoption.

Text Embedding 005

google/text-embedding-005

Google's latest text embedding model.

Text Embedding 004

google/text-embedding-004

Previous-gen Google embedding model.

Multilingual Embedding 002

google/text-multilingual-embedding-002

Google's multilingual text embedding model.

Voyage 3 Large

voyage/voyage-3-large

Voyage AI's most capable embedding model.

Voyage 3

voyage/voyage-3

Balanced performance and cost embedding model.

Voyage 3 Lite

voyage/voyage-3-lite

Lightweight embedding for cost-sensitive workloads.

Voyage Code 3

voyage/voyage-code-3

Code-specialized embedding for semantic code search.

Voyage Finance 2

voyage/voyage-finance-2

Finance-domain-optimized embedding model.

Voyage Law 2

voyage/voyage-law-2

Legal-domain-optimized embedding model.

Voyage Multilingual 2

voyage/voyage-multilingual-2

Multilingual embedding supporting diverse languages.

ERNIE 4.5 8K

baidu/ernie-4.5-8k

Baidu's flagship model with 8K context for general tasks.

ERNIE 4.5 128K

baidu/ernie-4.5-128k

Extended 128K context variant for long document analysis.

ERNIE 4.5 VL

baidu/ernie-4.5-vl

Vision-language variant with image understanding.

ERNIE 5.0 Thinking

baidu/ernie-5.0-thinking

Advanced reasoning with extended chain-of-thought.

ERNIE X1 Turbo

baidu/ernie-x1-turbo

Speed-optimized model for real-time applications.

Stable Diffusion 3

stability/stable-diffusion-3

Latest Stable Diffusion with improved text rendering and composition.

SD 3.5 Large

stability/sd-3.5-large

Large-scale Stable Diffusion for maximum image quality.

Stable Audio

stability/stable-audio

AI-powered audio and music generation from Stability AI.

Hunyuan Image V3

tencent/hunyuan-image-v3

Tencent's high-quality image generation model.

Hunyuan Part (3D)

tencent/hunyuan-part-3d

3D object generation and part-based modeling.

Magnum V4 72B

anthracite/magnum-v4-72b

Creative writing model with unrestricted generation.

Hermes 4 405B

nousresearch/hermes-4-405b

Massive 405B open-weight model with strong instruction following.

MythoMax-L2 13B

gryphe/mythomax-l2-13b

Popular merged model for creative and roleplay tasks.

Sharpen

topaz/sharpen

AI-powered image sharpening and enhancement.

Sharpen Generative

topaz/sharpen-generative

Generative AI sharpening for maximum detail recovery.

Reve Create Image

reve/create-image

AI image creation with artistic style control.

Reve Edit Image

reve/edit-image

AI-powered image editing and manipulation.

Reve Remix Image

reve/remix-image

Remix and transform images with AI.

ElevenLabs Multilingual V2

elevenlabs/multilingual-v2

High-quality multilingual text-to-speech in 29 languages.

ElevenLabs Turbo V2.5

elevenlabs/turbo-v2.5

Low-latency TTS optimized for real-time applications.

Eleven Music

elevenlabs/eleven-music

AI music generation with lyrics and instrumentals.

ElevenLabs V3 Alpha Voice

elevenlabs/v3-alpha-voice

Next-gen voice synthesis with emotional expression.

Aura 2 TTS

deepgram/aura-2-tts

Natural text-to-speech with low latency from Deepgram.

Octave 2

hume/octave-2

Emotionally expressive voice AI from Hume.

VibeVoice 1.5B

microsoft/vibevoice-1.5b

Compact TTS model with natural speech quality.

VibeVoice 7B

microsoft/vibevoice-7b

High-quality TTS with expressive voice generation.

TripoSR 3D

tripo/triposr-3d

Fast 3D object generation from single images.

Llama 3.3 70B (Groq)

groq/llama-3.3-70b

Ultra-fast inference on Groq hardware.

Llama 4 Maverick (Groq)

groq/llama-4-maverick

Fast open-source model on Groq.

DeepSeek R1 Distill (Groq)

groq/deepseek-r1-distill-llama-70b

Fast reasoning on Groq hardware.

Perplexity Sonar

perplexity/sonar

Web search with AI-generated answers.

Perplexity Sonar Pro

perplexity/sonar-pro

Thorough web search with citations.

Sonar Reasoning Pro

perplexity/sonar-reasoning-pro

Search-augmented reasoning.

Sonar Deep Research

perplexity/sonar-deep-research

Multi-query deep research.

DeepSeek R1 (Together)

together/deepseek-r1

Budget reasoning on Together.

Llama 4 Maverick (Together)

together/llama-4-maverick

Open-source model on Together.

Nemotron 70B

nvidia/llama-3.1-nemotron-70b

Reward-optimized model from Nvidia.

Vincony — Access the World's Best AI Models