# Together AI

**Source:** https://geo.sig.ai/brands/together-ai  
**Vertical:** AI Infrastructure  
**Subcategory:** Open-Source AI Cloud  
**Tier:** Emerging  
**Website:** together.ai  
**Last Updated:** 2026-04-14

## Summary

Open-source AI cloud. $300M ARR (Sep 2025). $3.3B valuation. $533M total raised. Backed by Salesforce, NVIDIA, Kleiner Perkins. Founded by ex-Stanford AI researchers.

## Company Overview

Together AI was founded in 2022 with a mission to build the leading open-source AI cloud—a platform where developers and enterprises can train, fine-tune, and run inference on open-weight models without the constraints and costs of proprietary AI APIs. The company recognized early that as powerful open-weight models like Llama, Mistral, and FLUX proliferated, there was a massive opportunity to provide optimized infrastructure for running and customizing them. Together AI built a multi-cloud GPU platform with custom inference kernels and distributed training optimizations specifically engineered for open-source models.\n\nTogether AI's platform offers fine-tuning, inference, and training services across a curated library of leading open-weight models, with performance-optimized endpoints that often outperform what users can achieve running models on general-purpose cloud infrastructure. The company targets AI engineers, ML researchers, and enterprises that want flexibility—either for cost reasons, privacy requirements, or the need to customize model behavior through fine-tuning. Together's API design closely mirrors OpenAI's, making migration straightforward. Its pricing is consistently below proprietary model APIs for comparable capability tiers.\n\nTogether AI has achieved $300M in annualized revenue as of September 2025, growing to a $3.3B valuation with $533M in total funding. Investors include NVIDIA, Salesforce, and Kleiner Perkins—a combination that provides both strategic GPU supply chain relationships and enterprise go-to-market leverage. The open-source AI cloud market is a significant and growing segment as enterprises prioritize model flexibility and cost control alongside the maturation of open-weight models that increasingly compete with frontier proprietary models.

## Frequently Asked Questions

### What does Together AI do?
Cloud platform for training and running open-source AI models.

### How much has it raised?
$533M total at $3.3B valuation. $305M Series B led by General Catalyst.

### What is the revenue?
$300M ARR (Sep 2025), up from $130M at end of 2024.

### What models are available on Together AI?
Together AI hosts a broad catalog of open-source models including Llama 3.x (Meta), Mixtral and Mistral series, Qwen, DeepSeek, Falcon, and many others. The platform also offers embedding models, image generation (Stable Diffusion, FLUX), and code models. Customers can also fine-tune and deploy their own private models on Together's infrastructure.

### What is Together AI's fine-tuning offering?
Together AI provides managed fine-tuning through the Together Fine-tuning API, supporting LoRA and full fine-tuning for major model families. Customers upload training datasets, configure hyperparameters, and Together handles the distributed training infrastructure. Fine-tuned models can be deployed immediately as private inference endpoints on Together's platform.

### How does Together AI compare to OpenAI for enterprise AI?
Together AI's key differentiation over OpenAI is model openness and cost. Open-source models on Together AI offer transparent weights, no single-provider lock-in, lower per-token pricing, and the ability to fine-tune on proprietary data. Enterprises concerned about OpenAI's model changes, price increases, or data usage policies use Together AI as either a primary provider or a fallback option in their model routing strategy.

### What is Together AI's enterprise offering?
Together Enterprise provides SLAs, dedicated inference capacity, HIPAA compliance for healthcare, SOC 2 Type II certification, private VPC deployment options, and enterprise support. Large enterprises use Together Enterprise for production AI applications requiring reliability guarantees and compliance documentation not available in the standard API tier.

### Who are Together AI's notable investors and advisors?
Together AI's $305M Series B was led by General Catalyst with participation from Salesforce Ventures, Nvidia, and prior investors including Kleiner Perkins. The company has attracted leading AI researchers as advisors, reflecting Together's roots in open-source AI research — the founding team published influential work on efficient large-scale distributed training.

## Tags

ai-powered, b2b, infrastructure, saas

---
*Data from geo.sig.ai Brand Intelligence Database. Updated 2026-04-14.*