Deepseek

DeepSeek-V4-Flash-0731

DeepSeek's official V4 Flash release, superseding the preview with substantially stronger agentic and coding performance, at 284B total / 13B active parameters and a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Google

Gemini-3.6-Flash

Available for Creator Tier+

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.6-Flash-Enterprise

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.5-Flash-Lite

Available for Creator Tier+

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Google

Gemini-3.5-Flash-Lite-Enterprise

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Qwen

Qwen3.8-Max-Preview

Alibaba Qwen 2.4T-parameter flagship preview with major gains in coding, full-stack development, data analysis, and office workflows; always uses thinking mode and supports long-horizon tool workflows and structured output

Moonshot AI

Kimi-K3

Moonshot AI's flagship model with a 1M-token context window, built for long-horizon agentic coding and tool use, with native vision input and always-on reasoning

Bytedance

Seedance-2.0

BytePlus Seedance 2.0 — unified multimodal video (text/image/video/audio refs, native audio). Billed on actual tokens; rate varies by resolution & reference video

Bytedance

Seedance-2.0-Fast

BytePlus Seedance 2.0 Fast — faster/cheaper variant (no 1080p). Billed on actual tokens

Bytedance

Seedance-2.0-Mini

BytePlus Seedance 2.0 Mini — cost-effective variant (no 1080p). Billed on actual tokens

Open AI

GPT-5.6-Sol

OpenAI GPT-5.6 Sol — flagship multimodal model in the GPT-5.6 series with top-tier reasoning and agentic coding

Open AI

GPT-5.6-Terra

OpenAI GPT-5.6 Terra — a balanced multimodal model in the GPT-5.6 series with strong reasoning, positioned between the flagship Sol and cost-efficient Luna tiers

xAI's flagship model released in July 2026, with frontier performance on coding, knowledge work, and STEM. 500K-token context window and multimodal text/image input

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Z.ai

GLM-5.2

Z.ai's open-source MoE model for long-context reasoning and agentic software engineering, with a 1M-token context window

Moonshot AI

Kimi-K2.7-Code

Moonshot AI's trillion-parameter open-source MoE coding model, built for long-horizon, high-efficiency agentic software engineering

Anthropic's most capable frontier model for demanding reasoning and long-horizon agentic work, with always-on adaptive thinking that adjusts reasoning depth based on task complexity

A million-token multimodal frontier model from MiniMax, built for long-horizon agentic work with native text/image/video understanding and tool calling

Qwen

Qwen3.6-35B-A3B

Qwen3.6 35B (3B-active) MoE with built-in NEXTN speculative decoding, reasoning and tool calling — fast low-cost generation

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Google

Gemini-3.5-Flash

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

xAI's reasoning model released in late April 2026, featuring always-on reasoning, a 1-million-token context window, and multimodal text/image input

Open AI

GPT-5.5

OpenAI's next-generation multimodal GPT-5.5 model with improved token efficiency for hard reasoning, coding, and agentic tasks

Deepseek

DeepSeek-V4-Pro

DeepSeek's next-generation flagship language model delivering state-of-the-art reasoning, coding, and agentic performance with a 1M context window

Deepseek

DeepSeek-V4-Flash

DeepSeek's fast, cost-efficient V4 variant optimized for high-throughput reasoning and coding with a 1M context window

Xiaomi's flagship text model tuned for complex software engineering, agentic workflows, and long-horizon tasks with a 1M context window

Xiaomi's natively omnimodal model accepting text, image, audio, and video inputs with a 1M context window for agentic and long-horizon reasoning

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Z.ai

GLM-5.1

Z.ai's open-source flagship MoE model built for agentic engineering, delivering state-of-the-art coding and long-horizon task performance that improves the longer it runs

Qwen

Qwen3.6-Plus

Alibaba's flagship hosted model with a 1M-token context window, excelling at agentic coding, multimodal reasoning, and long-horizon planning tasks

Google

Gemma-4-31B

Google DeepMind's open-weights, instruction-tuned multimodal model optimized for reasoning, coding, and agentic workflows

Mistral

Mistral-Small-4

A large (~119B parameter) dense language model designed for efficient, high-quality text generation and reasoning with a focus on strong performance-to-cost efficiency compared to similarly sized models

Google

Gemini-3.1-Flash-Lite-Preview

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

A next-generation agentic LLM with native interleaved thinking, built for complex real-world productivity tasks like coding, tool use, and multi-step reasoning — at a highly competitive price point

Z.ai

GLM-4.7-Flash

Available for Creator Tier+

A 30B Mixture-of-Experts language model optimized for efficient inference, strong coding performance, and long-context reasoning tasks

Qwen

Qwen3.5-Plus

Alibaba's native vision-language model with a 1M-token context window, hybrid MoE architecture, and built-in tool use, offering frontier-class multimodal reasoning at competitive pricing

Z.ai

GLM-5

Z.ai's most powerful model — a 744B-parameter (40B active) open-source MoE optimized for reasoning, coding, and agentic tasks

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

A lightweight, open-source text-to-text large language model optimized for coding, agent-oriented workflows, and instruction following, delivering faster, cleaner outputs and strong reasoning and problem-solving performance compared to its predecessor

Qwen

Qwen3-235B-A22B-Instruct-2507

A 235B-parameter MoE instruction-tuned language model for general, multilingual, and coding tasks

Z.ai

GLM-4.6

Flagship model from Z.ai with 200K token context, superior reasoning, advanced tool use, and coding excellence

Open AI

GPT-5.4

A frontier large language model optimized for complex professional tasks, combining strong reasoning, coding, and tool-use capabilities with very long context support

Open AI

GPT-5.3-Chat

A fast, general-purpose conversational LLM based on the GPT-5.3 architecture, optimized for high-quality dialogue, instruction following, and everyday chat applications

Open AI

GPT-5.3-Codex

A high-performance agentic coding model designed to autonomously write, debug, and manage complex software development tasks while collaborating interactively with developers

AI Agent Models

Cydonia-24B-v4.3

A 24B-parameter uncensored instruction-tuned model focused on creative writing with strong recall and prompt adherence

AI Agent Models

Cydonia-24B-v4.1

A 24B-parameter uncensored instruction-tuned model focused on creative writing with strong recall and prompt adherence

AI Agent Models

L3.3-70B-Euryale-v2.3

70B model focused on creative roleplay from Sao10k

AI Agent Models

L3.1-70B-Euryale-v2.2

70B model focused on creative roleplay from Sao10k

AI Agent Models

Sapphira-L3.3-70B-0.1

70B parameter language model optimized for storytelling, dialogue, and immersive role-play

AI Agent Models

MN-Violet-Lotus-12B

12B parameter model for creative, emotionally intelligent conversations

AI Agent Models

L3.3-MS-Nevoria-70B

70B narrative generation, LoRA tuning & story-rich outputs

Mistral

Mistral-Nemo-Instruct-2407

24B instruction model, long context, fewer errors

Google

Nano-Banana-2

Text-to-image and image+text-to-image with up to 14 reference images. Supports aspect_ratio, image_size, person_generation, output_mime_type, compression_quality, thinking_level.

Z-Image

Alibaba-Z-Image-Turbo

Alibaba Z-Image Turbo - Fast text-to-image with Chinese/English text rendering support.

Google

Nano-Banana-Pro-Edit

Premium AI-powered image editing with Gemini 3 Pro. Advanced editing capabilities with better quality, supports multiple resolution outputs (1K/2K/4K).

Google

Nano-Banana-Edit

AI-powered image editing with Gemini 2.5 Flash. Edit images using natural language instructions - remove backgrounds, change colors, apply styles, and more.

Bytedance

Bytedance-Seedream-5.0-Lite

Advanced text-to-image generation with web search, deep reasoning and smart editing capabilities. Supports up to 3K resolution and seed parameters.

Wan

Alibaba-Wan-2.6-T2V

Alibaba Wan 2.6 Flash - Fast text-to-video generation. Supports 2-15 seconds at 720p/1080p.

Wan

Alibaba-Wan-2.6-I2V

Alibaba Wan 2.6 I2V - Image-to-video generation. Supports 2-15 seconds at 720p/1080p.

Wan

Alibaba-Wan2.6-I2V-Flash

Alibaba Wan 2.6 I2V Flash - Fast lightweight I2V with 480P/720P/1080P, audio/silent pricing

Wan

Wan2.6-R2V

Alibaba Wan 2.6 R2V - Mixed image+video references (up to 5 combined)

Wan

Alibaba-Wan-2.6-R2V-Flash

Alibaba Wan 2.6 R2V Flash - Reference-to-video generation from 1-3 reference videos. Supports 5 or 10 seconds.

Google

Veo-3.1-Fast

Google Veo 3.1 Fast - Fast and economical text-to-video generation. Supports up to 8 seconds at 720p/1080p. Optional AI-generated audio (+50% cost).

Google

Veo-3.1-Fast-I2V

Google Veo 3.1 Fast Image-to-Video - Generate video from an input image. Supports up to 8 seconds at 720p/1080p.

Google

Veo-3.1

Google Veo 3.1 Standard - Balanced quality and speed for text-to-video generation. Supports up to 8 seconds at 720p/1080p.

Google

Veo-3.1-I2V

Google Veo 3.1 Standard Image-to-Video - Generate high-quality video from an input image.

Qwen

Qwen3-VL-Plus

Alibaba's multimodal vision-language API model that handles text, image, and video inputs with strong visual reasoning and agent capabilities

Qwen

Qwen3-VL-Flash

Alibaba's lightweight, cost-effective multimodal vision-language API model designed for fast inference on text, image, and video tasks with integrated thinking and non-thinking modes

MegaNova

Web Search

Real-time managed natural-language web search that returns ranked results, optionally enriched with full extracted page text in a single call—includes 50 free queries per user per day, with additional queries priced at $0.002 each

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

Google

Gemini-3.5-Flash-Enterprise

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite-Enterprise

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

Google

Gemini-3.1-Flash-Lite-Preview-Enterprise

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview-Enterprise

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Google

Gemini-3.1-Flash-Live

Gemini Live speech-to-speech model for real-time voice conversations — bidirectional audio streaming over the OpenAI-compatible /v1/realtime WebSocket

MegaNova

Manta-Mini-1.0

Lightweight tier optimized for speed and cost. (Free)

MegaNova

Manta-Flash-1.0

Balanced tier for general-purpose roleplay, good tradeoff between cost and latency. (Mid-tier)

MegaNova

Manta-Pro-1.0

Available for Creator Tier+

Premium tier with advanced reasoning and long context, best for complex roleplay.(Top-tier)

Deepseek

DeepSeek-V3.2

A high-performance open-weight large language model from DeepSeek optimized for strong reasoning, coding, and general chat capabilities with efficient inference

AI Agent Models

Cydonia-24B-v4.3

A 24B-parameter uncensored instruction-tuned model focused on creative writing with strong recall and prompt adherence

AI Agent Models

Cydonia-24B-v4.1

A 24B-parameter uncensored instruction-tuned model focused on creative writing with strong recall and prompt adherence

AI Agent Models

L3.3-70B-Euryale-v2.3

70B model focused on creative roleplay from Sao10k

AI Agent Models

L3.1-70B-Euryale-v2.2

70B model focused on creative roleplay from Sao10k

AI Agent Models

Sapphira-L3.3-70B-0.1

70B parameter language model optimized for storytelling, dialogue, and immersive role-play

AI Agent Models

MN-Violet-Lotus-12B

12B parameter model for creative, emotionally intelligent conversations

AI Agent Models

L3.3-MS-Nevoria-70B

70B narrative generation, LoRA tuning & story-rich outputs

AI Agent Models

L3-70B-Euryale-v2.1

70B tuned for storytelling, roleplay, creative writing

AI Agent Models

L3-8B-Stheno-v3.2

8B LLaMA-3 model for roleplay & assistant tasks

Deepseek

DeepSeek-V3.2-Exp

Experimental version with sparse-attention architecture for long-context efficiency

Deepseek

DeepSeek-V3-0324

High-performance open-source MoE language model optimized for reasoning, coding, and efficient text generation

Deepseek

DeepSeek-V3.1

Hybrid inference LLM with Think/Non-Think modes, 128K context, advanced agent

Deepseek

DeepSeek-R1-0528

An open-source next-generation reasoning-optimized language model with enhanced logic, math, and code performance, larger context handling, and improved function-calling capabilities

MegaNova

Manta-Mini-1.0

Lightweight tier optimized for speed and cost. (Free)

MegaNova

Manta-Flash-1.0

Balanced tier for general-purpose roleplay, good tradeoff between cost and latency. (Mid-tier)

MegaNova

Manta-Pro-1.0

Available for Creator Tier+

Premium tier with advanced reasoning and long context, best for complex roleplay.(Top-tier)

Deepseek

DeepSeek-V4-Flash-0731

DeepSeek's official V4 Flash release, superseding the preview with substantially stronger agentic and coding performance, at 284B total / 13B active parameters and a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Google

Gemini-3.6-Flash

Available for Creator Tier+

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.6-Flash-Enterprise

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.5-Flash-Lite

Available for Creator Tier+

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Google

Gemini-3.5-Flash-Lite-Enterprise

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Qwen

Qwen3.8-Max-Preview

Alibaba Qwen 2.4T-parameter flagship preview with major gains in coding, full-stack development, data analysis, and office workflows; always uses thinking mode and supports long-horizon tool workflows and structured output

Moonshot AI

Kimi-K3

Moonshot AI's flagship model with a 1M-token context window, built for long-horizon agentic coding and tool use, with native vision input and always-on reasoning

Open AI

GPT-5.6-Sol

OpenAI GPT-5.6 Sol — flagship multimodal model in the GPT-5.6 series with top-tier reasoning and agentic coding

Open AI

GPT-5.6-Terra

OpenAI GPT-5.6 Terra — a balanced multimodal model in the GPT-5.6 series with strong reasoning, positioned between the flagship Sol and cost-efficient Luna tiers

xAI's flagship model released in July 2026, with frontier performance on coding, knowledge work, and STEM. 500K-token context window and multimodal text/image input

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Z.ai

GLM-5.2

Z.ai's open-source MoE model for long-context reasoning and agentic software engineering, with a 1M-token context window

Moonshot AI

Kimi-K2.7-Code

Moonshot AI's trillion-parameter open-source MoE coding model, built for long-horizon, high-efficiency agentic software engineering

Anthropic's most capable frontier model for demanding reasoning and long-horizon agentic work, with always-on adaptive thinking that adjusts reasoning depth based on task complexity

A million-token multimodal frontier model from MiniMax, built for long-horizon agentic work with native text/image/video understanding and tool calling

Qwen

Qwen3.6-35B-A3B

Qwen3.6 35B (3B-active) MoE with built-in NEXTN speculative decoding, reasoning and tool calling — fast low-cost generation

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Google

Gemini-3.5-Flash

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

xAI's reasoning model released in late April 2026, featuring always-on reasoning, a 1-million-token context window, and multimodal text/image input

Open AI

GPT-5.5

OpenAI's next-generation multimodal GPT-5.5 model with improved token efficiency for hard reasoning, coding, and agentic tasks

Deepseek

DeepSeek-V4-Pro

DeepSeek's next-generation flagship language model delivering state-of-the-art reasoning, coding, and agentic performance with a 1M context window

Deepseek

DeepSeek-V4-Flash

DeepSeek's fast, cost-efficient V4 variant optimized for high-throughput reasoning and coding with a 1M context window

Xiaomi's flagship text model tuned for complex software engineering, agentic workflows, and long-horizon tasks with a 1M context window

Xiaomi's natively omnimodal model accepting text, image, audio, and video inputs with a 1M context window for agentic and long-horizon reasoning

Moonshot AI

Kimi-K2.6

An open-weight 1T-parameter MoE multimodal model built for long-horizon agentic coding, with a 256K context window, native vision input, and scaling up to 300 parallel sub-agents

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Z.ai

GLM-5.1

Z.ai's open-source flagship MoE model built for agentic engineering, delivering state-of-the-art coding and long-horizon task performance that improves the longer it runs

Qwen

Qwen3.6-Plus

Alibaba's flagship hosted model with a 1M-token context window, excelling at agentic coding, multimodal reasoning, and long-horizon planning tasks

Google

Gemma-4-31B

Google DeepMind's open-weights, instruction-tuned multimodal model optimized for reasoning, coding, and agentic workflows

AI Agent Models

MiMo-V2-Omni

Xiaomi's omni-modal agent model that natively understands text, images, video, and audio in a unified architecture, designed for complex real-world multimodal tasks

Mistral

Mistral-Small-4

A large (~119B parameter) dense language model designed for efficient, high-quality text generation and reasoning with a focus on strong performance-to-cost efficiency compared to similarly sized models

Google

Gemini-3.1-Flash-Lite-Preview

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

A trillion-parameter text-only flagship agent model from Xiaomi, built for complex coding, long-horizon planning, and multi-step tool use with a 1M context window

A next-generation agentic LLM with native interleaved thinking, built for complex real-world productivity tasks like coding, tool use, and multi-step reasoning — at a highly competitive price point

Z.ai

GLM-4.7-Flash

Available for Creator Tier+

A 30B Mixture-of-Experts language model optimized for efficient inference, strong coding performance, and long-context reasoning tasks

Qwen

Qwen3.5-Plus

Alibaba's native vision-language model with a 1M-token context window, hybrid MoE architecture, and built-in tool use, offering frontier-class multimodal reasoning at competitive pricing

MiniMax-M2.5 is a cost-efficient frontier model that achieves SOTA in coding, agentic tool use, and search tasks through extensive reinforcement learning in real-world environments

Z.ai

GLM-5

Z.ai's most powerful model — a 744B-parameter (40B active) open-source MoE optimized for reasoning, coding, and agentic tasks

Moonshot AI

Kimi-K2.5

An open-source native multimodal AI model that combines vision and language understanding with advanced agent-oriented capabilities, including self-directed agent swarm and visual coding, for versatile reasoning and tool use

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Deepseek

DeepSeek-V3.2

A high-performance open-weight large language model from DeepSeek optimized for strong reasoning, coding, and general chat capabilities with efficient inference

A lightweight, open-source text-to-text large language model optimized for coding, agent-oriented workflows, and instruction following, delivering faster, cleaner outputs and strong reasoning and problem-solving performance compared to its predecessor

Z.ai

GLM-4.7

An open-weights MoE chat model from Z.ai optimized for strong reasoning and tool-using agents

Z.ai

GLM-4.7-Enterprise

An open-weights MoE chat model from Z.ai optimized for strong reasoning and tool-using agents

Qwen

Qwen3-235B-A22B-Instruct-2507

A 235B-parameter MoE instruction-tuned language model for general, multilingual, and coding tasks

Moonshot AI

Kimi-K2-Thinking

1T-parameter reasoning-focused language model designed for complex, multi-step problem solving and tool use

Z.ai

GLM-4.6

Flagship model from Z.ai with 200K token context, superior reasoning, advanced tool use, and coding excellence

Google

Gemini-3-Flash-Preview

Google’s agentic workhorse model, bringing near Pro agentic, coding and multimodal intelligence, with more balanced cost and speed

Google

Gemini-2.5-Pro

Strongest Gemini, 1M context, great for code & knowledge

Google

Gemini-2.5-Flash

Fast reasoning with improved latency & accuracy

Google

Gemini-2.5-Flash-Lite

Ultra-low latency Gemini variant

Open AI

GPT-5.4

A frontier large language model optimized for complex professional tasks, combining strong reasoning, coding, and tool-use capabilities with very long context support

Open AI

GPT-5.3-Chat

A fast, general-purpose conversational LLM based on the GPT-5.3 architecture, optimized for high-quality dialogue, instruction following, and everyday chat applications

Open AI

GPT-5.3-Codex

A high-performance agentic coding model designed to autonomously write, debug, and manage complex software development tasks while collaborating interactively with developers

Open AI

GPT-5.2

OpenAI’s flagship multimodal (text + image) GPT-5.2 model for top-tier coding and agentic tasks, producing text output with the highest reasoning capability in the GPT-5 family

Open AI

GPT-5.1

OpenAI’s multimodal (text + image) GPT-5 model for strong coding, reasoning, and agentic tasks across domains, producing text outputs

Open AI

GPT-5

OpenAI’s multimodal (text + image) GPT-5 model for strong coding, reasoning, and agentic tasks across domains, producing text outputs

Open AI

GPT-5-Mini

OpenAI’s faster, more cost-efficient GPT-5 model for well-defined tasks, supporting text and image input with text output

Open AI

GPT-5-Nano

OpenAI’s fastest, most cost-efficient GPT-5 model for lightweight tasks like summarization and classification, supporting text and image input with text output

Open AI

GPT-4o-mini

Compact GPT-4 Omni, supports text & image inputs

AI Agent Models

Cydonia-24B-v4.3

A 24B-parameter uncensored instruction-tuned model focused on creative writing with strong recall and prompt adherence

AI Agent Models

Cydonia-24B-v4.1

A 24B-parameter uncensored instruction-tuned model focused on creative writing with strong recall and prompt adherence

AI Agent Models

L3.3-70B-Euryale-v2.3

70B model focused on creative roleplay from Sao10k

AI Agent Models

L3.1-70B-Euryale-v2.2

70B model focused on creative roleplay from Sao10k

AI Agent Models

Sapphira-L3.3-70B-0.1

70B parameter language model optimized for storytelling, dialogue, and immersive role-play

AI Agent Models

MN-Violet-Lotus-12B

12B parameter model for creative, emotionally intelligent conversations

AI Agent Models

L3.3-MS-Nevoria-70B

70B narrative generation, LoRA tuning & story-rich outputs

Mistral

Mistral-Nemo-Instruct-2407

24B instruction model, long context, fewer errors

Mistral

Mistral-Small-3.2-24B-Instruct-2506

24B instruction model, long context, fewer errors

AI Agent Models

L3-70B-Euryale-v2.1

70B tuned for storytelling, roleplay, creative writing

AI Agent Models

L3-8B-Stheno-v3.2

8B LLaMA-3 model for roleplay & assistant tasks

Deepseek

DeepSeek-V3.2-Exp

Experimental version with sparse-attention architecture for long-context efficiency

Deepseek

DeepSeek-V3-0324

High-performance open-source MoE language model optimized for reasoning, coding, and efficient text generation

Deepseek

DeepSeek-V3-0324-Enterprise

High-performance open-source MoE language model optimized for reasoning, coding, and efficient text generation

Deepseek

DeepSeek-V3.1

Hybrid inference LLM with Think/Non-Think modes, 128K context, advanced agent

Deepseek

DeepSeek-V3.1-Enterprise

Hybrid inference LLM with Think/Non-Think modes, 128K context, advanced agent

Deepseek

DeepSeek-R1-0528

An open-source next-generation reasoning-optimized language model with enhanced logic, math, and code performance, larger context handling, and improved function-calling capabilities

Llama

Llama3.3-70B

Multilingual 70B delivering 405B-level performance

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

Google

Gemini-3.5-Flash-Enterprise

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite-Enterprise

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

Google

Gemini-3.1-Flash-Lite-Preview-Enterprise

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview-Enterprise

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Google

Gemini-3-Flash-Preview-Enterprise

Google’s agentic workhorse model, bringing near Pro agentic, coding and multimodal intelligence, with more balanced cost and speed

Google

Gemini-2.5-Pro-Enterprise

Strongest Gemini, 1M context, great for code & knowledge

Google

Gemini-2.5-Flash-Enterprise

Fast reasoning with improved latency & accuracy

Google

Gemini-2.5-Flash-Lite-Enterprise

Ultra-low latency Gemini variant

Deepseek

DeepSeek-V4-Flash-0731

DeepSeek's official V4 Flash release, superseding the preview with substantially stronger agentic and coding performance, at 284B total / 13B active parameters and a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Google

Gemini-3.6-Flash

Available for Creator Tier+

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.6-Flash-Enterprise

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.5-Flash-Lite

Available for Creator Tier+

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Google

Gemini-3.5-Flash-Lite-Enterprise

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Qwen

Qwen3.8-Max-Preview

Alibaba Qwen 2.4T-parameter flagship preview with major gains in coding, full-stack development, data analysis, and office workflows; always uses thinking mode and supports long-horizon tool workflows and structured output

Moonshot AI

Kimi-K3

Moonshot AI's flagship model with a 1M-token context window, built for long-horizon agentic coding and tool use, with native vision input and always-on reasoning

Bytedance

Seedance-2.0

BytePlus Seedance 2.0 — unified multimodal video (text/image/video/audio refs, native audio). Billed on actual tokens; rate varies by resolution & reference video

Bytedance

Seedance-2.0-Fast

BytePlus Seedance 2.0 Fast — faster/cheaper variant (no 1080p). Billed on actual tokens

Bytedance

Seedance-2.0-Mini

BytePlus Seedance 2.0 Mini — cost-effective variant (no 1080p). Billed on actual tokens

Open AI

GPT-5.6-Sol

OpenAI GPT-5.6 Sol — flagship multimodal model in the GPT-5.6 series with top-tier reasoning and agentic coding

Open AI

GPT-5.6-Terra

OpenAI GPT-5.6 Terra — a balanced multimodal model in the GPT-5.6 series with strong reasoning, positioned between the flagship Sol and cost-efficient Luna tiers

xAI's flagship model released in July 2026, with frontier performance on coding, knowledge work, and STEM. 500K-token context window and multimodal text/image input

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Z.ai

GLM-5.2

Z.ai's open-source MoE model for long-context reasoning and agentic software engineering, with a 1M-token context window

Moonshot AI

Kimi-K2.7-Code

Moonshot AI's trillion-parameter open-source MoE coding model, built for long-horizon, high-efficiency agentic software engineering

Anthropic's most capable frontier model for demanding reasoning and long-horizon agentic work, with always-on adaptive thinking that adjusts reasoning depth based on task complexity

A million-token multimodal frontier model from MiniMax, built for long-horizon agentic work with native text/image/video understanding and tool calling

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Google

Gemini-3.5-Flash

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

xAI's reasoning model released in late April 2026, featuring always-on reasoning, a 1-million-token context window, and multimodal text/image input

Open AI

GPT-5.5

OpenAI's next-generation multimodal GPT-5.5 model with improved token efficiency for hard reasoning, coding, and agentic tasks

Deepseek

DeepSeek-V4-Pro

DeepSeek's next-generation flagship language model delivering state-of-the-art reasoning, coding, and agentic performance with a 1M context window

Deepseek

DeepSeek-V4-Flash

DeepSeek's fast, cost-efficient V4 variant optimized for high-throughput reasoning and coding with a 1M context window

Xiaomi's flagship text model tuned for complex software engineering, agentic workflows, and long-horizon tasks with a 1M context window

Xiaomi's natively omnimodal model accepting text, image, audio, and video inputs with a 1M context window for agentic and long-horizon reasoning

Moonshot AI

Kimi-K2.6

An open-weight 1T-parameter MoE multimodal model built for long-horizon agentic coding, with a 256K context window, native vision input, and scaling up to 300 parallel sub-agents

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Z.ai

GLM-5.1

Z.ai's open-source flagship MoE model built for agentic engineering, delivering state-of-the-art coding and long-horizon task performance that improves the longer it runs

Qwen

Qwen3.6-Plus

Alibaba's flagship hosted model with a 1M-token context window, excelling at agentic coding, multimodal reasoning, and long-horizon planning tasks

Google

Gemma-4-31B

Google DeepMind's open-weights, instruction-tuned multimodal model optimized for reasoning, coding, and agentic workflows

Mistral

Mistral-Small-4

A large (~119B parameter) dense language model designed for efficient, high-quality text generation and reasoning with a focus on strong performance-to-cost efficiency compared to similarly sized models

Google

Gemini-3.1-Flash-Lite-Preview

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

A next-generation agentic LLM with native interleaved thinking, built for complex real-world productivity tasks like coding, tool use, and multi-step reasoning — at a highly competitive price point

Z.ai

GLM-4.7-Flash

Available for Creator Tier+

A 30B Mixture-of-Experts language model optimized for efficient inference, strong coding performance, and long-context reasoning tasks

Qwen

Qwen3.5-Plus

Alibaba's native vision-language model with a 1M-token context window, hybrid MoE architecture, and built-in tool use, offering frontier-class multimodal reasoning at competitive pricing

MiniMax-M2.5 is a cost-efficient frontier model that achieves SOTA in coding, agentic tool use, and search tasks through extensive reinforcement learning in real-world environments

Z.ai

GLM-5

Z.ai's most powerful model — a 744B-parameter (40B active) open-source MoE optimized for reasoning, coding, and agentic tasks

Moonshot AI

Kimi-K2.5

An open-source native multimodal AI model that combines vision and language understanding with advanced agent-oriented capabilities, including self-directed agent swarm and visual coding, for versatile reasoning and tool use

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

Deepseek

DeepSeek-V3.2

A high-performance open-weight large language model from DeepSeek optimized for strong reasoning, coding, and general chat capabilities with efficient inference

Z.ai

GLM-4.7

An open-weights MoE chat model from Z.ai optimized for strong reasoning and tool-using agents

Deepseek

DeepSeek-V3-0324-Enterprise

High-performance open-source MoE language model optimized for reasoning, coding, and efficient text generation

Deepseek

DeepSeek-V3.1-Enterprise

Hybrid inference LLM with Think/Non-Think modes, 128K context, advanced agent

Google

Nano-Banana-2

Text-to-image and image+text-to-image with up to 14 reference images. Supports aspect_ratio, image_size, person_generation, output_mime_type, compression_quality, thinking_level.

Z-Image

Alibaba-Z-Image-Turbo

Alibaba Z-Image Turbo - Fast text-to-image with Chinese/English text rendering support.

Bytedance

Bytedance-Seedream-5.0-Lite

Advanced text-to-image generation with web search, deep reasoning and smart editing capabilities. Supports up to 3K resolution and seed parameters.

Bytedance

Bytedance-Seedream-4.5

The latest Bytedance image model, delivering better editing consistency, improved multi-image fusion, finer detail control, natural small text and faces, and harmonious, aesthetic visuals

Wan

Alibaba-Wan-2.6-T2V

Alibaba Wan 2.6 Flash - Fast text-to-video generation. Supports 2-15 seconds at 720p/1080p.

Wan

Alibaba-Wan-2.6-I2V

Alibaba Wan 2.6 I2V - Image-to-video generation. Supports 2-15 seconds at 720p/1080p.

Wan

Alibaba-Wan2.6-I2V-Flash

Alibaba Wan 2.6 I2V Flash - Fast lightweight I2V with 480P/720P/1080P, audio/silent pricing

Wan

Wan2.6-R2V

Alibaba Wan 2.6 R2V - Mixed image+video references (up to 5 combined)

Wan

Alibaba-Wan-2.6-R2V-Flash

Alibaba Wan 2.6 R2V Flash - Reference-to-video generation from 1-3 reference videos. Supports 5 or 10 seconds.

Qwen

Qwen3-VL-Plus

Alibaba's multimodal vision-language API model that handles text, image, and video inputs with strong visual reasoning and agent capabilities

Qwen

Qwen3-VL-Flash

Alibaba's lightweight, cost-effective multimodal vision-language API model designed for fast inference on text, image, and video tasks with integrated thinking and non-thinking modes

MegaNova

Web Search

Real-time managed natural-language web search that returns ranked results, optionally enriched with full extracted page text in a single call—includes 50 free queries per user per day, with additional queries priced at $0.002 each

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

Google

Gemini-3.5-Flash-Enterprise

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite-Enterprise

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

Google

Gemini-3.1-Flash-Lite-Preview-Enterprise

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview-Enterprise

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Google

Gemini-3.1-Flash-Live

Gemini Live speech-to-speech model for real-time voice conversations — bidirectional audio streaming over the OpenAI-compatible /v1/realtime WebSocket

Deepseek

DeepSeek-V4-Flash-0731

DeepSeek's official V4 Flash release, superseding the preview with substantially stronger agentic and coding performance, at 284B total / 13B active parameters and a 1M-token context window

Deepseek

DeepSeek-V4-Pro

DeepSeek's next-generation flagship language model delivering state-of-the-art reasoning, coding, and agentic performance with a 1M context window

Deepseek

DeepSeek-V4-Flash

DeepSeek's fast, cost-efficient V4 variant optimized for high-throughput reasoning and coding with a 1M context window

Deepseek

DeepSeek-V3.2

A high-performance open-weight large language model from DeepSeek optimized for strong reasoning, coding, and general chat capabilities with efficient inference

Deepseek

DeepSeek-V3.2-Exp

Experimental version with sparse-attention architecture for long-context efficiency

Deepseek

DeepSeek-V3-0324

High-performance open-source MoE language model optimized for reasoning, coding, and efficient text generation

Deepseek

DeepSeek-V3-0324-Enterprise

High-performance open-source MoE language model optimized for reasoning, coding, and efficient text generation

Deepseek

DeepSeek-V3.1

Hybrid inference LLM with Think/Non-Think modes, 128K context, advanced agent

Deepseek

DeepSeek-V3.1-Enterprise

Hybrid inference LLM with Think/Non-Think modes, 128K context, advanced agent

Deepseek

DeepSeek-R1-0528

An open-source next-generation reasoning-optimized language model with enhanced logic, math, and code performance, larger context handling, and improved function-calling capabilities

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Google

Gemini-3.6-Flash

Available for Creator Tier+

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.6-Flash-Enterprise

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.5-Flash-Lite

Available for Creator Tier+

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Google

Gemini-3.5-Flash-Lite-Enterprise

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Moonshot AI

Kimi-K3

Moonshot AI's flagship model with a 1M-token context window, built for long-horizon agentic coding and tool use, with native vision input and always-on reasoning

Open AI

GPT-5.6-Sol

OpenAI GPT-5.6 Sol — flagship multimodal model in the GPT-5.6 series with top-tier reasoning and agentic coding

Open AI

GPT-5.6-Terra

OpenAI GPT-5.6 Terra — a balanced multimodal model in the GPT-5.6 series with strong reasoning, positioned between the flagship Sol and cost-efficient Luna tiers

xAI's flagship model released in July 2026, with frontier performance on coding, knowledge work, and STEM. 500K-token context window and multimodal text/image input

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Moonshot AI

Kimi-K2.7-Code

Moonshot AI's trillion-parameter open-source MoE coding model, built for long-horizon, high-efficiency agentic software engineering

Anthropic's most capable frontier model for demanding reasoning and long-horizon agentic work, with always-on adaptive thinking that adjusts reasoning depth based on task complexity

A million-token multimodal frontier model from MiniMax, built for long-horizon agentic work with native text/image/video understanding and tool calling

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Google

Gemini-3.5-Flash

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

xAI's reasoning model released in late April 2026, featuring always-on reasoning, a 1-million-token context window, and multimodal text/image input

Open AI

GPT-5.5

OpenAI's next-generation multimodal GPT-5.5 model with improved token efficiency for hard reasoning, coding, and agentic tasks

Xiaomi's natively omnimodal model accepting text, image, audio, and video inputs with a 1M context window for agentic and long-horizon reasoning

Moonshot AI

Kimi-K2.6

An open-weight 1T-parameter MoE multimodal model built for long-horizon agentic coding, with a 256K context window, native vision input, and scaling up to 300 parallel sub-agents

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Qwen

Qwen3.6-Plus

Alibaba's flagship hosted model with a 1M-token context window, excelling at agentic coding, multimodal reasoning, and long-horizon planning tasks

Google

Gemma-4-31B

Google DeepMind's open-weights, instruction-tuned multimodal model optimized for reasoning, coding, and agentic workflows

AI Agent Models

MiMo-V2-Omni

Xiaomi's omni-modal agent model that natively understands text, images, video, and audio in a unified architecture, designed for complex real-world multimodal tasks

Mistral

Mistral-Small-4

A large (~119B parameter) dense language model designed for efficient, high-quality text generation and reasoning with a focus on strong performance-to-cost efficiency compared to similarly sized models

Google

Gemini-3.1-Flash-Lite-Preview

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Qwen

Qwen3.5-Plus

Alibaba's native vision-language model with a 1M-token context window, hybrid MoE architecture, and built-in tool use, offering frontier-class multimodal reasoning at competitive pricing

Moonshot AI

Kimi-K2.5

An open-source native multimodal AI model that combines vision and language understanding with advanced agent-oriented capabilities, including self-directed agent swarm and visual coding, for versatile reasoning and tool use

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Google

Gemini-3-Flash-Preview

Google’s agentic workhorse model, bringing near Pro agentic, coding and multimodal intelligence, with more balanced cost and speed

Google

Gemini-2.5-Pro

Strongest Gemini, 1M context, great for code & knowledge

Google

Gemini-2.5-Flash

Fast reasoning with improved latency & accuracy

Google

Gemini-2.5-Flash-Lite

Ultra-low latency Gemini variant

Open AI

GPT-5.4

A frontier large language model optimized for complex professional tasks, combining strong reasoning, coding, and tool-use capabilities with very long context support

Open AI

GPT-5.3-Chat

A fast, general-purpose conversational LLM based on the GPT-5.3 architecture, optimized for high-quality dialogue, instruction following, and everyday chat applications

Open AI

GPT-5.3-Codex

A high-performance agentic coding model designed to autonomously write, debug, and manage complex software development tasks while collaborating interactively with developers

Open AI

GPT-5.2

OpenAI’s flagship multimodal (text + image) GPT-5.2 model for top-tier coding and agentic tasks, producing text output with the highest reasoning capability in the GPT-5 family

Open AI

GPT-5.1

OpenAI’s multimodal (text + image) GPT-5 model for strong coding, reasoning, and agentic tasks across domains, producing text outputs

Open AI

GPT-5

OpenAI’s multimodal (text + image) GPT-5 model for strong coding, reasoning, and agentic tasks across domains, producing text outputs

Open AI

GPT-5-Mini

OpenAI’s faster, more cost-efficient GPT-5 model for well-defined tasks, supporting text and image input with text output

Open AI

GPT-5-Nano

OpenAI’s fastest, most cost-efficient GPT-5 model for lightweight tasks like summarization and classification, supporting text and image input with text output

Open AI

GPT-4o-mini

Compact GPT-4 Omni, supports text & image inputs

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

Google

Gemini-3.5-Flash-Enterprise

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite-Enterprise

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

Google

Gemini-3.1-Flash-Lite-Preview-Enterprise

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview-Enterprise

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Google

Gemini-3-Flash-Preview-Enterprise

Google’s agentic workhorse model, bringing near Pro agentic, coding and multimodal intelligence, with more balanced cost and speed

Google

Gemini-2.5-Pro-Enterprise

Strongest Gemini, 1M context, great for code & knowledge

Google

Gemini-2.5-Flash-Enterprise

Fast reasoning with improved latency & accuracy

Google

Gemini-2.5-Flash-Lite-Enterprise

Ultra-low latency Gemini variant

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Anthropic's most capable model for complex agentic coding and enterprise work, delivering a step-change in deep reasoning and long-horizon autonomy over Claude Opus 4.8, with a 1M-token context window

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most capable frontier model for demanding reasoning and long-horizon agentic work, with always-on adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

Anthropic's most capable generally available model, excelling at agentic coding, long-running tasks, and complex professional work with adaptive thinking that adjusts reasoning depth based on task complexity

Anthropic's most intelligent model, excelling at complex reasoning, coding, analysis, and multi-step tasks

An advanced flagship model delivering top-tier reasoning, deep analysis, and long-context performance for the most demanding enterprise and research workloads

Top-tier, large-scale reasoning model designed for complex analysis, long-context understanding, and high-accuracy enterprise-grade tasks

Anthropic's most capable mid-tier model, delivering near-Opus-level intelligence across coding, computer use, agentic workflows, and long-context reasoning with a 1M-token context window

A fast, cost-efficient model optimized for everyday reasoning, coding, and high-quality conversational tasks

A lightweight, low-latency model built for rapid responses, simple reasoning, and high-throughput applications

MegaNova

Manta-Mini-1.0

Lightweight tier optimized for speed and cost. (Free)

MegaNova

Manta-Flash-1.0

Balanced tier for general-purpose roleplay, good tradeoff between cost and latency. (Mid-tier)

MegaNova

Manta-Pro-1.0

Available for Creator Tier+

Premium tier with advanced reasoning and long context, best for complex roleplay.(Top-tier)

Google

Gemini-3.6-Flash

Available for Creator Tier+

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.5-Flash-Lite

Available for Creator Tier+

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Z.ai

GLM-4.7-Flash

Available for Creator Tier+

A 30B Mixture-of-Experts language model optimized for efficient inference, strong coding performance, and long-context reasoning tasks

AI Agent Models

Sapphira-L3.3-70B-0.1

70B parameter language model optimized for storytelling, dialogue, and immersive role-play

AI Agent Models

MN-Violet-Lotus-12B

12B parameter model for creative, emotionally intelligent conversations

AI Agent Models

L3.3-MS-Nevoria-70B

70B narrative generation, LoRA tuning & story-rich outputs

Mistral

Mistral-Small-3.2-24B-Instruct-2506

24B instruction model, long context, fewer errors

AI Agent Models

L3-70B-Euryale-v2.1

70B tuned for storytelling, roleplay, creative writing

AI Agent Models

L3-8B-Stheno-v3.2

8B LLaMA-3 model for roleplay & assistant tasks

Google

Gemini-3.6-Flash

Available for Creator Tier+

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.6-Flash-Enterprise

Google's latest Flash model — upgraded multimodal reasoning, coding and agentic performance at Flash speed and cost, with a 1M-token context window

Google

Gemini-3.5-Flash-Lite

Available for Creator Tier+

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Google

Gemini-3.5-Flash-Lite-Enterprise

Google's most cost-efficient Gemini 3.5 model — low-latency multimodal inference tuned for high-volume classification, extraction and translation workloads

Google

Gemini-3.5-Flash

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

Google

Gemini-3.1-Flash-Lite-Preview

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Google

Gemini-3-Flash-Preview

Google’s agentic workhorse model, bringing near Pro agentic, coding and multimodal intelligence, with more balanced cost and speed

Google

Gemini-2.5-Pro

Strongest Gemini, 1M context, great for code & knowledge

Google

Gemini-2.5-Flash

Fast reasoning with improved latency & accuracy

Google

Gemini-2.5-Flash-Lite

Ultra-low latency Gemini variant

Google

Nano-Banana-2

Text-to-image and image+text-to-image with up to 14 reference images. Supports aspect_ratio, image_size, person_generation, output_mime_type, compression_quality, thinking_level.

Google

Nano-Banana-Pro-Edit

Premium AI-powered image editing with Gemini 3 Pro. Advanced editing capabilities with better quality, supports multiple resolution outputs (1K/2K/4K).

Google

Nano-Banana-Pro

High-quality image generation with better text rendering, character consistency, and advanced composition. Powered by Gemini 3 Pro Image.

Google

Nano-Banana-Edit

AI-powered image editing with Gemini 2.5 Flash. Edit images using natural language instructions - remove backgrounds, change colors, apply styles, and more.

Google

Nano-Banana

Gemini 2.5 Flash Image - High-quality image generation with multiple aspect ratios

Google

Gemini-2.5-Flash-TTS

A low-latency, cost-efficient text-to-speech model that converts text into expressive single or multi-speaker audio with controllable tone, style, and pacing

Google

Gemini-2.5-Pro-TTS

A high-quality text-to-speech model designed to generate natural, expressive audio with fine-grained control over tone, pacing, and emotion for narration, assistants, and conversational voice applications

Google

Gemini-3.5-Flash-Enterprise

Latest Gemini 3.5 Flash — fast multimodal reasoning with 1M context

Google

Gemini-3.1-Flash-Lite-Enterprise

Google's most cost-efficient, low-latency multimodal model in the Gemini 3 series, optimized for high-volume, latency-sensitive tasks like translation, classification, and simple data extraction with support for text, image, video, audio, and PDF inputs

Google

Gemini-3.1-Flash-Lite-Preview-Enterprise

Google’s fastest and most cost-efficient Gemini 3 model, optimized for high-throughput tasks and scalable real-time applications

Google

Gemini-3.1-Pro-Preview-Enterprise

Google's frontier reasoning model that builds on the Gemini 3 Pro series with enhanced thinking capabilities, improved token efficiency, and stronger performance in software engineering and agentic workflows, featuring a 1M-token context window

Google

Gemini-3-Flash-Preview-Enterprise

Google’s agentic workhorse model, bringing near Pro agentic, coding and multimodal intelligence, with more balanced cost and speed

Google

Gemini-2.5-Pro-Enterprise

Strongest Gemini, 1M context, great for code & knowledge

Google

Gemini-2.5-Flash-Enterprise

Fast reasoning with improved latency & accuracy

Google

Gemini-2.5-Flash-Lite-Enterprise

Ultra-low latency Gemini variant

Google

Gemini-3.1-Flash-Live

Gemini Live speech-to-speech model for real-time voice conversations — bidirectional audio streaming over the OpenAI-compatible /v1/realtime WebSocket

Qwen

Qwen3.8-Max-Preview

Alibaba Qwen 2.4T-parameter flagship preview with major gains in coding, full-stack development, data analysis, and office workflows; always uses thinking mode and supports long-horizon tool workflows and structured output