Explore / AI Models

The model catalog,
exactly as documented.

Every model below is copied from the ChatBoxAI models documentation, with its exact chatboxai/ model ID and tier. Free accounts get a smaller model set with request caps; paid accounts with credits unlock the full catalog.

Start for FreeRead the documentation ↗
Listed here
23 named models
Namespace
chatboxai/
Auto selection
"model": "auto"

01 / Directory

Find a model

23 of 23 models

Chat

Everyday conversation, writing, extraction and normal queries.

  • Llama 3.3 70Bchatboxai/llama-3.3-70b

    Best overall open-source model. Excellent for general chat, creative writing, and robust instruction following.

    Free
  • Gemini 2.5 Flashchatboxai/gemini-2.5-flash

    Extremely fast, next-generation multimodal model. Best for high-speed text and vision tasks.

    Free
  • Qwen3 32Bchatboxai/qwen3-32b

    Strong coding and logic reasoning. Ideal for general-purpose programming queries.

    Free
  • Qwen 3.6 Plus Previewchatboxai/qwen-3.6-plus

    Advanced multi-turn conversation and coding logic.

    Free
  • Mistral Small 3.1chatboxai/mistral-small-3.1

    Fast and lightweight with native function calling support.

    Free
  • GLM 4.5 Airchatboxai/glm-4.5-air

    Built-in chain-of-thought processing for structured problem solving.

    Free
  • Llama 3.1 8B Turbochatboxai/llama-3.1-8b

    Ultra-fast, high-throughput model. Best for simple queries and API fallback.

    Free
  • Gemma 3 27Bchatboxai/gemma-3-27b

    Highly capable, balanced text and vision processing.

    Free
  • Claude Opus 4.7chatboxai/claude-opus-4.7

    Flagship reasoning. Best for highly complex problem solving, deep analysis, and high-fidelity content generation.

    Paid
  • Claude 4.5 Sonnetchatboxai/claude-4.5-sonnet

    Perfect balance of speed and top-tier intelligence. Ideal for versatile everyday tasks.

    Paid
  • Gemini 2.5 Prochatboxai/gemini-2.5-pro

    Premium performance with massive context (up to 2M tokens). Best for analyzing entire codebases or huge document sets.

    Paid
  • Qwen3 Coder 480Bchatboxai/qwen3-coder

    Heavy-duty coding. Best for building complex applications and debugging huge codebases.

    Paid
  • GPT-OSS 120Bchatboxai/gpt-oss-120b

    Top-tier open-source synthesis and architectural design.

    Paid

Deep Research

Multi-step reasoning, agents, tool use and web research workflows.

  • Kimi K2chatboxai/kimi-k2

    Highly optimized for agentic workflows and tool usage. Capable of handling deep web searches and synthesizing information autonomously.

    Free
  • NVIDIA Nemotron 3 Superchatboxai/nemotron-3-super

    Massive 1M context window combined with deep agentic reasoning. Best for orchestrating complex research workflows over massive document libraries.

    Paid
  • Hermes 3 405Bchatboxai/hermes-3-405b

    Frontier reasoning and strict tool-use. Best for handling multi-step logic puzzles and autonomous agents.

    Paid
  • Trinity Large Previewchatboxai/trinity-large

    Massive 400B MoE model. Best for expansive, complex persona adoption during agentic interactions.

    Paid

Image

Used with the POST /api/v1/images/generations endpoint.

  • FLUX Schnellchatboxai/image-flux-schnell

    High-speed image generation. Ideal for rapid visual prototyping and quick concept art.

    Free
  • FLUX Devchatboxai/image-flux-dev

    Ultra-high-quality image generation. Best for production-ready artwork, hyper-realistic photos, and detailed visuals.

    Paid
  • Stable Diffusion 3.5chatboxai/image-stable-diffusion-3.5

    Highly flexible generation. Best for complex prompt adherence and highly stylized artistic rendering.

    Paid
  • Stable Diffusion XLchatboxai/image-sdxl

    Cinematic and photographic visuals. Best for highly realistic character portraits and environments.

    Paid

Audio

Used with the POST /api/v1/audio/speech endpoint for text-to-speech.

  • TTS-1tts-1

    Standard text-to-speech for everyday voice generation and basic app accessibility features.

    Free
  • TTS-1 HDtts-1-hd

    High-definition, low-noise audio generation. Best for professional voiceovers, podcasts, and media production.

    Paid

The docs also note additional lightweight free-tier models: llama-3.2-3b, qwen3-next-80b, minimax-m2.5, nemotron-3-nano. The catalog changes over time — the live models documentation ↗ is the authoritative current list.

02 / Choosing

A short selection guide

  1. Everyday chat & writing

    Start with a Free chat model such as Llama 3.3 70B or Gemini 2.5 Flash.

  2. Coding

    Qwen3 32B on Free; Qwen3 Coder 480B on Paid for heavy-duty work.

  3. Research & agents

    Kimi K2 on Free; Nemotron 3 Super or Hermes 3 405B on Paid.

  4. Not sure?

    Pass "model": "auto" and ChatBoxAI picks an available model based on latency and availability.

Switch models without rewriting your app.

Try ChatBoxAI FreeSee the workspace ↓