> ## Documentation Index > Fetch the complete documentation index at: https://docs.together.ai/llms.txt > Use this file to discover all available pages before exploring further. # Overview > Run, train, and serve open-source AI models on Together AI. export const ModelGrid = () => { const modelGroups = [{ title: "Chat models:", link: "/docs/serverless/models#chat-models", hasViewAll: true, items: [{ name: "MiniMax M2.5", icon: "/images/intro/minimax.svg", description: "228.7B parameter model with 80.2% SWE-Bench, optimized for agentic coding tasks.", link: "https://www.together.ai/models/minimax-m2-5" }, { name: "Kimi K2.5", icon: "/images/intro/kimi.svg", description: "State-of-the-art multimodal model with advanced reasoning, vision, and coding capabilities.", link: "https://www.together.ai/models/kimi-k2-5" }, { name: "GLM-5", icon: "/images/intro/glm.svg", description: "Zhipu's latest large language model with strong reasoning and instruction following capabilities.", link: "https://www.together.ai/models/glm-5" }, { name: "Qwen3.5 397B", icon: "/images/intro/qwen.svg", description: "397B parameter (17B activated) MoE model with state-of-the-art reasoning, coding, and multilingual capabilities.", link: "https://www.together.ai/models/qwen3-5-397b-a17b" }, { name: "DeepSeek V3.1", icon: "/images/intro/deepseek.svg", description: "671B parameters (37B activated), 128K context, hybrid thinking/non-thinking modes, advanced tool calling, agent capabilities", link: "https://www.together.ai/models/deepseek-v3-1" }, { name: "GPT OSS 120B", icon: "/images/intro/gpt.svg", description: "OpenAI's open-source reasoning model with chain-of-thought, Apache 2.0 licensed.", link: "https://www.together.ai/models/gpt-oss-120b" }, { name: "GPT OSS 20B", icon: "/images/intro/gpt.svg", description: "Compact open-source reasoning model by OpenAI, Apache 2.0 licensed.", link: "https://www.together.ai/models/gpt-oss-20b" }] }, { title: "Audio models:", link: "/docs/serverless/models#audio-models", hasViewAll: true, items: [{ name: "MiniMax Speech 2.6", icon: "/images/intro/minimax.svg", description: "Sub-250ms latency text-to-speech with 40+ language support.", link: "https://www.together.ai/models/minimax-speech-2-6-turbo" }, { name: "Cartesia Sonic 3", icon: "/images/intro/cartesia.svg", description: "Next-generation low-latency, ultra-realistic voice model with improved naturalness and expressiveness.", link: "https://www.together.ai/models/cartesia-sonic-3" }, { name: "Orpheus 3B", icon: "/images/intro/orpheus.svg", description: "Expressive text-to-speech model with natural prosody and emotion control.", link: "https://www.together.ai/models/orpheus-3b" }, { name: "Kokoro", icon: "/images/intro/kokoro.svg", description: "Lightweight 82M parameter text-to-speech model with high-quality voice synthesis.", link: "https://www.together.ai/models/kokoro" }, { name: "Whisper Large v3", icon: "/images/intro/gpt.svg", description: "High-performance speech-to-text model delivering transcription 15x faster than OpenAI with support for 1GB+ files, 50+ languages.", link: "https://www.together.ai/models/openai-whisper-large-v3" }, { name: "Parakeet TDT 0.6B", icon: "/images/intro/nvidia.svg", description: "NVIDIA's efficient speech-to-text model with strong accuracy across multiple languages.", link: "https://www.together.ai/models/parakeet-tdt-0-6b" }] }, { title: "Image models:", link: "/docs/serverless/models#image-models", hasViewAll: true, items: [{ name: "Nano Banana Pro", icon: "/images/intro/google.svg", iconSmall: true, description: "Google's Gemini 3 Pro image generation with state-of-the-art text rendering and 2K/4K output.", link: "https://www.together.ai/models/nano-banana-pro" }, { name: "FLUX.2 [pro]", icon: "/images/intro/flux.svg", description: "Latest premium image generation model by Black Forest Labs.", link: "https://www.together.ai/models/flux-2-pro" }, { name: "FLUX.1 [schnell]", icon: "/images/intro/flux.svg", description: "Fastest available endpoint for the SOTA open-source image generation model by Black Forest Labs.", link: "https://www.together.ai/models/flux-1-schnell" }] }, { title: "Vision models:", link: "/docs/serverless/models#vision-models", hasViewAll: true, items: [{ name: "Qwen3.5 397B", icon: "/images/intro/qwen.svg", description: "397B parameter MoE model with native vision capabilities for advanced visual reasoning and understanding.", link: "https://www.together.ai/models/qwen3-5-397b-a17b" }, { name: "Kimi K2.5", icon: "/images/intro/kimi.svg", description: "Multimodal model excelling at visual understanding, document analysis, and image reasoning.", link: "https://www.together.ai/models/kimi-k2-5" }, { name: "Qwen3.5 9B", icon: "/images/intro/qwen.svg", description: "Efficient 9B multimodal model with native vision, video understanding, and agentic capabilities.", link: "https://www.together.ai/models/qwen3-5-9b" }] }]; const getGridStyle = index => { const styles = [{ gridRow: "span 4" }, { gridRow: "span 4" }, { gridRow: "span 2" }, { gridRow: "span 2" }]; return styles[index] || ({}); }; return <>
{modelGroups.map((group, index) => { const models = group.items; return

{group.title}

{group.hasViewAll &&

View all models

}
{models.map((item, i) => {group.hasViewAll &&

{item.name}

} )}
{!group.hasViewAll && }
; })}
; }; export const WideCtaCard = ({title, description, iconUrl, href}) => { const cardContent =
{iconUrl && }

{title}

{description}

; return href ? {cardContent} : cardContent; }; export const CtaCard = ({title, description, border = true, coloredIcon = false, iconUrl, darkIconUrl, href}) => { const cardContent =
{iconUrl ? <> {darkIconUrl && } : }

{title}

{description}

; return href ? {cardContent} : cardContent; }; export const GridCards = ({children}) => { return
{children}
; }; export const QuickstartWrapper = ({children}) => { return <>

Developer quickstart

Copy this snippet to get started with our inference API. See the{" "} full quickstart {" "} for more details.

{children}
; }; export const Quickstart = ({}) => { return

python

typescript

curL

from {" "} together{" "} import {" "} Together
client{" "} = {" "} Together()

completion{" "} = {" "} client.chat.completions.create(
{" "} model = "Qwen/Qwen3.5-9B" ,
{" "} messages = [{"{"} "role" :{" "} "user" ,{" "} "content" :{" "} "What are the top 3 things to do in New York?" {"}"}],)

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

123456789

Developer quickstart

Copy this snippet to get started with our inference API. See the{" "} full quickstart for more details.

; }; export const SubHeading = ({heading, description}) => { return

{heading}

{description &&

{description}

}
; }; ```python Python theme={null} from together import Together client = Together() completion = client.chat.completions.create( model="openai/gpt-oss-20b", messages=[{"role": "user", "content": "What are the top 3 things to do in New York?"}], ) print(completion.choices[0].message.content) ``` ```typescript TypeScript theme={null} import Together from 'together-ai'; const together = new Together(); const completion = await together.chat.completions.create({ model: 'openai/gpt-oss-20b', messages: [{ role: 'user', content: 'Top 3 things to do in New York?' }], }); console.log(completion.choices[0].message.content); ``` ```bash cURL theme={null} curl -X POST "https://api.together.ai/v1/chat/completions" \ -H "Authorization: Bearer $TOGETHER_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "openai/gpt-oss-20b", "messages": [ {"role": "user", "content": "What are the top 3 things to do in New York?"} ] }' ``` Run leading open-source AI models with our OpenAI-compatible API. Fine-tune models on your own data and deploy them for inference. Spin up H100 and B200 clusters with attached storage for training or large batch jobs.
{" "} {" "} {" "} {" "} {" "} {" "} {" "} {" "} {" "}