KrixIsHere
Model catalogue

Every model, one workspace

21 chat models and 11 image & video models — pick any of them in the dashboard. The same catalogue powers this page and the chat.

21 chat8 image3 videoOpen dashboard
Token basics

What are AI tokens?

Tokens are the small pieces of text an AI model reads and writes. A token can be a whole short word, part of a longer word, punctuation, or a number. Models use tokens to measure request size, response length, context limits, and often cost.

Input tokens

Your prompt, attachments, instructions, and relevant chat history.

Output tokens

The response generated by the model.

Context window

The total amount of input and output a model can consider at once.

As a rough English-language guide, 100 tokens is often around 70-80 words. Exact counts vary by model, language, formatting, and tokenizer.

Token estimator

Type or paste text to estimate its input token count.

11

Est. tokens

9

Words

50

Characters

This is a local estimate. The final API response may report a different count because each model provider tokenizes text differently.

Chat & reasoning models

Grouped by provider. Models marked with a wrench can call tools (image generation, functions).

OpenAI

OpenAI

· 5
OpenAI
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

GPT-5 mini

Fast flagship — vision & tools

gpt-5-mini

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.64K outMax output tokens the model can return.
OpenAI
Tool calling — this model can generate images and call functions on its own.ProPro — uses paid credits.

GPT-5.4

Most powerful OpenAI — reasoning

gpt-5-4

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.64K outMax output tokens the model can return.
OpenAI
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

GPT-5.4 mini

Balanced speed & quality

gpt-5-4-mini

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.64K outMax output tokens the model can return.
OpenAI
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

GPT-5 nano

Fastest & cheapest

gpt-5-nano

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.32K outMax output tokens the model can return.
OpenAI
FreeFree — use free credits.

GPT-OSS 120B

OpenAI's open-source 120B model for powerful NPC dialogue and game logic.

gpt-oss-120b

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.32K outMax output tokens the model can return.
Claude

Anthropic

· 4
Claude
Tool calling — this model can generate images and call functions on its own.ProPro — uses paid credits.

Claude Opus 4.8

Most powerful — agents & reasoning

claude-opus-4-8

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.32K outMax output tokens the model can return.
Claude
Tool calling — this model can generate images and call functions on its own.ProPro — uses paid credits.

Claude Opus 4.7

Premium reasoning and agentic tasks.

claude-opus-4-7

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.16K outMax output tokens the model can return.
Claude
Tool calling — this model can generate images and call functions on its own.ProPro — uses paid credits.

Claude Sonnet 4.6

Balanced reasoning & speed

claude-sonnet-4-6

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.32K outMax output tokens the model can return.
Claude
Tool calling — this model can generate images and call functions on its own.ProPro — uses paid credits.

Claude Haiku 4.5

Fast structured writing

claude-haiku-4-5

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.16K outMax output tokens the model can return.
Gemini

Google

· 3
Gemini
Tool calling — this model can generate images and call functions on its own.ProPro — uses paid credits.

Gemini 3.1 Pro

Frontier multimodal reasoning

gemini-3-1-pro-preview

1M ctxContext window — total tokens the model can consider at once.1M inMax input tokens per request.64K outMax output tokens the model can return.
Gemini
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

Gemini 3 Flash

Enhanced multimodal — vision & tools

gemini-3-flash

1M ctxContext window — total tokens the model can consider at once.1M inMax input tokens per request.8K outMax output tokens the model can return.
Gemini
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

Gemini 2.5 Flash Lite

Fast & cheap

gemini-2-5-flash-lite

1M ctxContext window — total tokens the model can consider at once.1M inMax input tokens per request.8K outMax output tokens the model can return.
Grok

xAI

· 2
Grok
FreeFree — use free credits.

Grok 4.3

Reasoning with live knowledge

grok-4-3

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.16K outMax output tokens the model can return.
Grok
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

Grok 4.2

Vision & tools

grok-4-2

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.16K outMax output tokens the model can return.
DeepSeek

DeepSeek

· 2
DeepSeek
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

DeepSeek V4 Flash

Fast reasoning & tools

deepseek-v4-flash

128K ctxContext window — total tokens the model can consider at once.128K inMax input tokens per request.8K outMax output tokens the model can return.
DeepSeek
FreeFree — use free credits.

DeepSeek V4 Pro

Deep reasoning

deepseek-v4-pro

128K ctxContext window — total tokens the model can consider at once.128K inMax input tokens per request.8K outMax output tokens the model can return.
Kimi

Moonshot

· 1
Kimi
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

Kimi K2.6

Vision, reasoning & tools

kimi-k2-6

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.16K outMax output tokens the model can return.
Qwen

Alibaba

· 1
Qwen
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

Qwen Large

Vision, reasoning & tools

qwen-large

256K ctxContext window — total tokens the model can consider at once.256K inMax input tokens per request.16K outMax output tokens the model can return.
Zhipu

Z.ai

· 1
Zhipu
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

GLM 5.1

Reasoning & tools

glm-5-1

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.16K outMax output tokens the model can return.
Minimax

MiniMax

· 2
Minimax
Tool calling — this model can generate images and call functions on its own.FreeFree — use free credits.

MiniMax M2.5

Reasoning & tools

minimax-m2-5

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.16K outMax output tokens the model can return.
Minimax
FreeFree — use free credits.

MiniMax M3.0

Reasoning & tools

minimax-m3

200K ctxContext window — total tokens the model can consider at once.200K inMax input tokens per request.16K outMax output tokens the model can return.

Image & video models

Used automatically when you ask the agent to create visuals, or with the /image command.

image

Flux Schnell

Ultra-fast Flux generation. Best speed-to-quality ratio.

flux-schnell0.5 credits / image
image

Z-Image Turbo

Turbo-speed 6B Flux with 2x upscaling for vivid output.

z-image-turbo1 credit / image
OpenAIimage

GPT Image 1 Mini

OpenAI image generation and editing.

gpt-image-1-mini5 credits / image
OpenAIimage

GPT Image 2

High-quality image generation and editing with vision support at 720p.

gpt-image-215 credits / image
OpenAIimage

GPT Image 1.5

Enhanced quality image generation and editing with vision support.

gpt-image-1-510 credits / image
image

Wan 2.7 Image

Vision-capable image generation with image input support.

wan-2-7-image10 credits / image
Qwenimage

Qwen Image Plus

High-quality image generation from Alibaba.

qwen-image-plus7 credits / image
Geminiimage

Nano Banana 2

Google Gemini image generation and editing with reference image fusion.

nano-banana-225 credits / image
video

Wan 2.2

Fast video generation with image input support.

wan-2-250 credits / video
video

LTX-2

Video generation for text prompts.

ltx-240 credits / video
video

Amazon Nova Reel

High-quality video generation from text or image input.

nova-reel70 credits / video