Together AI
NativeAI NativePrivateUpdated November 15, 2026The open-source model infrastructure company combining research credibility with commercial traction.
Try it free →- Website
- together.ai
- Founded
- 2022
- CEO
- Vipul Ved Prakash
- HQ
- San Francisco, CA, United States
- Stage
- Series C
- Revenue
- $100M–$500M
- Total Funding
- $1.3B
- Employees
- 201-500
Executive Summary
Updated Nov 12, 2026
What they do
- Cloud platform for running, fine-tuning and training open-weight AI models
- Serverless, batch and dedicated inference APIs plus self-serve and reserved NVIDIA GPU clusters
- Research-led performance stack including FlashAttention lineage, kernels and the new Together Compiler
Why it matters
- Raised $800M Series C at $8.3B valuation led by Aramco Ventures, July 2026
- Annual bookings reportedly topped $1.15B, serving Cursor, Cognition, Decagon and Salesforce
- Multiyear capacity deals with IBM Cloud ($240M) and Rackspace ($380M) expand GPU supply
Best fit
- Enterprises moving production workloads from closed APIs to open-weight models
- AI-native product teams needing fast, low-cost inference on the newest open models
- Teams wanting inference, fine-tuning, GPU clusters and agent sandboxes on one bill
Watch-outs
- Capital-intensive GPU business depends on third-party data centers and NVIDIA supply
- Competitive market with Fireworks, Baseten, CoreWeave, Groq and hyperscaler model services
- Status trackers logged short major outages in September and October 2026
Products & Offerings
Together Inference
Serverless, batch and dedicated endpoints for hundreds of open-weight language, code, image, audio and video models, with day-zero access to new releases.
Together GPU Clusters (Accelerated Compute)
Self-serve instant clusters and reserved NVIDIA GPU clusters scaling to thousands of GPUs, optimized with the Together Kernel Collection.
Together Fine-Tuning
Managed fine-tuning service supporting LoRA, full fine-tuning and preference optimization on open models, billed per training token.
Together Code Sandbox
Fast, secure isolated VM sandboxes at scale for AI agents and development environments that execute code.
Together Compiler
Uses agents to continually rewrite and optimize the GPU code behind served open models, raising inference throughput as traffic changes.
Recent Releases
- Polymorphic InferenceProductNewsNov 2026
- Together CompilerProductNewsNov 2026
- Together Post-Training ServiceProductNewsOct 2026
- AuroraModelHugging FaceFeb 2026
Latest News
CASRAI · 1 day ago
PositiveHoodline · 3 days ago
PositiveData Center Dynamics · 4 days ago
PositiveTipRanks · 6 days ago
PositiveTradingView · 6 days ago
PositiveUpdated Nov 12, 2026
Together AI is one of the most established independent platforms for running open-weight models in production. It has strong research credentials, a broad product range, large customers and fresh funding, which makes it a default shortlist candidate for buyers moving off closed-model APIs. Before you commit, check its uptime commitments (SLAs), where your data will be processed and stored, and how firm its capacity is for your target GPUs, and run it head-to-head against Fireworks, Baseten and your hyperscaler's model service.
Signal Provenance
Latest observation date and source for each signal category
| Signal Category | Last Updated | Source |
|---|---|---|
Funding Signals Last funding round, Funding stage | Nov 14, 2026 | Cited web research |
Hiring Velocity Headcount, Headcount growth | Nov 15, 2026 | Research agents (estimate) |
Technical Output Research papers, GitHub activity | Nov 15, 2026 | Hugging Face Hub + news feed + research agents |
News Sentiment News mentions, News sentiment | Nov 15, 2026 | Verified news feed |
Partnership Signals Enterprise partnerships, Distribution partners | Nov 15, 2026 | Research agents (estimate) |
Revenue Indicators ARR estimate, Revenue growth | Nov 15, 2026 | Research agents (estimate) |
Leadership Signals CEO tenure, Founder-led | Nov 15, 2026 | Research agents (estimate) |
Competitive Position Category rank, Market share | Nov 15, 2026 | Research agents (estimate) |
Why this Bluescore?
6 dimensions, each scored 0–10 against all 250 companies. Each adds its weight × score to the Bluescore — see methodology
Together AI's Bluescore of 69.5 rests most on Product Depth (8.7, top 1%) and Talent & Research (9.3, top 2%). It is above the AI250 median on all six dimensions; its lowest is Customer Adoption (5.5, 69th percentile).
In plain English
Together AI's Bluescore of 69.5 out of 100 is driven mainly by Talent & Research, where it scores 9.3 of 10 at the 98th percentile, and Product Depth, where it scores 8.7 of 10 at the 99th percentile. The biggest drag is Business Model & Revenue, which scores 5.1 of 10 and adds only 10.6 of 21 points, leaving 10.4 points on the table, the largest gap of any dimension. Improving that revenue score would likely move the overall ranking more than any other change, since it carries a 21% weight and is currently well below the company's strengths.
Written by AI from the scores below and checked against them; updated nightly.
Traction Signals
Last Round
Series C
Risks
Updated Nov 12, 2026