Daniox

One AI platform. Chat, API, and intelligent model access.

A place to think. A platform to build. Choose how you use Daniox.

Chat

Free

Chat with AI models through Daniox.

Write, explore ideas, and work through questions. Keep your conversations in one place.

Open Chat

API

For builders

Build applications with Daniox's unified AI API.

Manage keys, choose a model, and track token usage with clear, usage-based pricing.

Open API

Free Chat · no API key requiredAPI billed separately by usageExplore supported models

01The platform

One place to use AI. One place to build with it.

Daniox brings conversations, model access, and developer controls together—without making you manage each integration yourself.

A focused Chat workspace

Choose a Chat model, revisit saved conversations, and work with streamed answers. Free to use, with service limits.

Your developer workspace

Create and rotate API keys, set permitted models and limits, and keep credentials out of your frontend.

A layer above the models

Call a published model directly or use Daniox Auto for policy-based selection through the same chat-completions endpoint.

Understand recorded usage

See token counts, charges, and recorded request details in the developer platform. API billing is separate from free Chat.

Controls, not guesswork

Use scoped keys and account limits. Model access depends on configuration and permissions; the catalog is not an uptime guarantee.

A familiar request shape

Use OpenAI-compatible chat completions with JSON or streaming responses. Advanced features vary by model and endpoint.

Go to Chat →Go to the API platform →Read documentation →

02Model access

Choose a model. Keep the same integration.

Explore Daniox's published API catalog. Chat has its own smaller model selection; API access depends on runtime availability and your key permissions.

Find a model

13 of 13 published models

Daniox Smart

daniox-smart

General-purpose text conversations.

Context not publishedChat / Streaming

$0.16 in · $0.475 out / 1M tokens (USD)

Daniox Flash

daniox-flash

A lower-priced option for text conversations.

Context not publishedChat / Streaming

$0.035 in · $0.145 out / 1M tokens (USD)

Daniox Vision

daniox-vision

Image input through supported API requests; text-only in Chat.

Context not publishedChat / Vision

$0.035 in · $0.145 out / 1M tokens (USD)

DeepSeek V4 Flash

deepseek-v4-flash-0731

DeepSeek-family chat completions.

262K contextChat / Streaming

$0.154 in · $0.308 out / 1M tokens (USD)

DeepSeek V4.1 Flash

deepseek-v4.1-flash

DeepSeek-family chat completions.

262K contextChat / Streaming

$0.24 in · $0.96 out / 1M tokens (USD)

GLM 5.3 Flash

glm-5.3-flash

GLM-family chat completions.

262K contextChat / Streaming

$0.09 in · $0.3 out / 1M tokens (USD)

Agents-A1

Agents-A1

Published Agents-A1 chat model.

262K contextChat / Streaming

$0.048 in · $0.36 out / 1M tokens (USD)

Devstral 2 123B

Devstral-2-123B-Instruct-2512-int4-AutoRound

Devstral-family chat completions.

128K contextChat / Streaming

$0.07 in · $0.36 out / 1M tokens (USD)

Gemma 4 31B

gemma-4-31B-it-fp8

Gemma-family chat completions.

262K contextChat / Streaming

$0.022 in · $0.07 out / 1M tokens (USD)

GPT-OSS 20B

gpt-oss-20b

GPT-OSS-family chat completions.

20K contextChat / Streaming

$0.014 in · $0.05 out / 1M tokens (USD)

Qwen3-Coder-Next

Qwen3-Coder-Next-FP8

Qwen Coder-family chat completions.

256K contextChat / Streaming

$0.036 in · $0.18 out / 1M tokens (USD)

Qwen3.6 35B

Qwen3.6-35B-A3B-FP8

Qwen-family chat completions.

262K contextChat / Streaming

$0.024 in · $0.18 out / 1M tokens (USD)

Qwen 3.8 27B

Qwen3.8-27B-FP8

Qwen-family chat completions.

262K contextChat / Streaming

$0.09 in · $0.64 out / 1M tokens (USD)

Prefer policy-based selection?

Use daniox-auto in the API. $0.012 in · $0.04 out / 1M.

Meet Daniox Auto →

Context windows and capabilities shown here are published metadata, not live checks. Unknown limits are not unlimited. Check runtime metadata in the platform and validate advanced features for your chosen model. Read model documentation.

03Daniox Auto

One model ID. Policy-based selection.

Auto selects from eligible models using task signals and routing policies. Or choose an explicit model when consistency and control matter most.

daniox-auto

Available through the public chat-completions API

Routing scores are configured estimates, not measured benchmarks. Auto does not promise the best answer, the fastest response, or uninterrupted recovery after streaming starts. It is not currently a Chat or API Playground model choice.

Read the task signals
Prompt patterns and estimated length inform a heuristic task classification.
Apply routing policies
Selection uses the eligible catalog, configured preferences, known context limits, cost estimates, and available health signals.
Send to a selected model
Daniox handles the upstream connection. A retry or fallback may be attempted before a successful stream begins.

04Developer quickstart

From API key to your first response.

One base URL for supported chat-completion models. Compatibility is specific to each endpoint and model—not the entire OpenAI API.

Create a scoped API key
Manage model access, key expiry, and request limits in the developer platform. Keep your key on the server.
Use a familiar interface
Send chat-completion requests with OpenAI client libraries or HTTP. Use JSON responses or SSE streaming where supported.
Inspect your usage
Review recorded requests, token usage, and charges. The API and API Playground are metered separately from free Chat.

Open API →Read documentation

Node.js with ES modules: npm install openai. Set DANIOX_API_KEY in your server environment.

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.DANIOX_API_KEY,
  baseURL: "https://api.daniox.tech/v1",
});

const completion = await client.chat.completions.create({
  model: "daniox-auto",
  messages: [{ role: "user", content: "Hello, Daniox." }],
});

console.log(completion.choices[0].message.content);

Python: pip install openai. Set DANIOX_API_KEY in your environment.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["DANIOX_API_KEY"],
    base_url="https://api.daniox.tech/v1",
)

completion = client.chat.completions.create(
    model="daniox-auto",
    messages=[{"role": "user", "content": "Hello, Daniox."}],
)

print(completion.choices[0].message.content)

Use a Bash-compatible shell with curl. Set DANIOX_API_KEY in your environment.

curl https://api.daniox.tech/v1/chat/completions \
  -H "Authorization: Bearer $DANIOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "daniox-auto",
    "messages": [{ "role": "user", "content": "Hello, Daniox." }]
  }'

Run these examples on your server with a funded API account and an eligible key. Never embed secrets in browser code.

05Pricing

Free Chat. Clear API pricing.

Chat is free, with service limits. The API and API Playground are billed separately by token usage. Published API rates below are in USD per 1 million tokens.

Open free Chat Open API

Daniox Auto

daniox-auto

Our lowest published API rate: below all 13 individual models below. Use this model ID with the public chat-completions API to let Daniox select an eligible model.

Input / 1M tokens: $0.012
Output / 1M tokens: $0.04

Individual model rates · USD per 1M tokens · input and output billed separately

Daniox models

Daniox models: published rates in US dollars per 1M tokens
Model	Input	Output
Daniox Smartdaniox-smart	$0.16	$0.475
Daniox Flashdaniox-flash	$0.035	$0.145
Daniox Visiondaniox-vision	$0.035	$0.145

Ecosystem models

Ecosystem models: published rates in US dollars per 1M tokens
Model	Input	Output
DeepSeek V4 Flashdeepseek-v4-flash-0731	$0.154	$0.308
DeepSeek V4.1 Flashdeepseek-v4.1-flash	$0.24	$0.96
GLM 5.3 Flashglm-5.3-flash	$0.09	$0.3
Agents-A1Agents-A1	$0.048	$0.36
Devstral 2 123BDevstral-2-123B-Instruct-2512-int4-AutoRound	$0.07	$0.36
Gemma 4 31Bgemma-4-31B-it-fp8	$0.022	$0.07
GPT-OSS 20Bgpt-oss-20b	$0.014	$0.05
Qwen3-Coder-NextQwen3-Coder-Next-FP8	$0.036	$0.18
Qwen3.6 35BQwen3.6-35B-A3B-FP8	$0.024	$0.18
Qwen 3.8 27BQwen3.8-27B-FP8	$0.09	$0.64

API use requires a funded account and an eligible key. Model access and limits depend on your account and key configuration. Usage pages show recorded token counts and charges; publication does not guarantee live availability. These rates do not apply to free Chat.

How Auto selects a model →

06Good to know

Clear answers before you start.

What is the difference between Chat and API?

Chat is a free conversational workspace. The API is for integrating models into your own software and is billed by token usage. The API Playground uses API billing; it is not the free Chat workspace.

Do I need an API key or credits to use Chat?

No. Sign in to Chat with your Daniox account. Chat does not require an API key or a funded API balance. Service and model limits still apply.

Are all API models also in Chat?

No. Chat currently offers Daniox Smart, Daniox Flash, and Daniox Vision as text-conversation choices. The developer API has a wider published catalog. Image upload, Auto selection, tools, and web search are not Chat controls today.

How does Daniox Auto choose a model?

Auto selects from eligible models using task signals and routing policies. Selection considers configured model characteristics, estimated tokens and cost, and available health signals. It is not a guarantee of the fastest or highest-quality answer. You can also specify an exact public model ID.

Can I use an existing OpenAI client?

OpenAI-compatible chat-completions clients can use the Daniox base URL and a Daniox key. Compatibility is scoped to supported endpoints and model features; it does not mean every OpenAI API is implemented.

Where can I check access and availability?

The authenticated model catalog shows runtime metadata. GET /v1/models lists enabled public model IDs. Key permissions and account configuration still determine access. Check the status page for service notices; a catalog entry is not a live health check.

Start a conversation. Build something new.

Two ways into Daniox. Choose yours.

Open Chat Open API

One AI platform. Chat, API, and intelligent model access.

Chat

API

One place to use AI. One place to build with it.

A focused Chat workspace

Your developer workspace

A layer above the models

Understand recorded usage

Controls, not guesswork

A familiar request shape

Choose a model. Keep the same integration.

Daniox Smart

Daniox Flash

Daniox Vision

DeepSeek V4 Flash

DeepSeek V4.1 Flash

GLM 5.3 Flash

Agents-A1

Devstral 2 123B

Gemma 4 31B

GPT-OSS 20B

Qwen3-Coder-Next

Qwen3.6 35B

Qwen 3.8 27B

Prefer policy-based selection?

One model ID. Policy-based selection.

Read the task signals

Apply routing policies

Send to a selected model

From API key to your first response.

1.Create a scoped API key

2.Use a familiar interface

3.Inspect your usage

Free Chat. Clear API pricing.

Daniox Auto

Daniox models

Ecosystem models

Clear answers before you start.

Start a conversation. Build something new.

Create a scoped API key

Use a familiar interface

Inspect your usage