Just 24 hours after launching @concentrateai's HIPAA offering:
In addition to large AI companies, we've already seen signups from:
- Small psychology and therapy practices
- Independent law firms that can't justify expensive legal AI software
- Healthcare practices looking for alternatives to clunky, single-purpose AI tools
- Professionals using local AI apps like @getopenwork and self-hosted @OpenWebUI.
Many of these businesses have been unable to use mainstream AI tools with sensitive data, or have been stuck with expensive, specialized software.
Now they can use GPT, Claude, Gemini, and open-source models through one gateway, with a BAA covering eligible inference.
The coolest part is seeing how creatively people are using AI once they can choose their own models and tools.
@concentrateai has a $5K VC startup program!
We have dozens of @ycombinator, @a16z@speedrun , @DraperVC and more startups on the platform. BAAs are needed not just for healthcare but legaltech, insurtech, fintech, etc.!
Reach out for up to $5K in free credits.
Healthcare and life sciences companies deserve to innovate, earn a margin, and increase enterprise value while running AI in production - and now they can!
Introducing Concentrate AI (@concentrateai), the first HIPAA-compliant AI gateway for healthcare and life sciences.
One API and one universal BAA with us for HIPAA-covered inference.
We handle provider agreements, inference, and billing for you, with spend controls and fallbacks built in. Customers are saving over 50% with lower-cost open-source models.
SOC 2 Type 2. HIPAA enabled. Built for personal and patient data (PII/PHI) across healthcare, pharma, legaltech, insurtech, and fintech.
Now generally available. Thanks to early customers: Pocket, Kinstead Health, Hummingbird Healthcare, and many more.
See what your AI setup is missing and book a demo: https://t.co/tMDaCuUVhj
Meta's Muse Ember is now live on Concentrate AI.
@Meta and @alexandr_wang's smallest and most cost-efficient reasoning model yet:
- $0.10/M input, $0.50/M output
- 1M context, 131K max output
- Outperforms Haiku and Luna on Meta-cited benchmarks
Built for high-volume coding agents, tool calls, and document processing.
We're thrilled to announce our partnership with @relace_ai, now live on @concentrateai!
It's been so fun working with @pfactorialz and @EBorgnia.
One of the biggest names in low-cost AI inference just joined the only no-fee AI gateway.
Relace has quickly become one of the most popular inference providers across AI gateways, processing trillions of tokens with some of the most competitive prices for open-weight models.
Now you can access their DeepSeek, GLM, Kimi, and more through Concentrate, with ZERO fees.
DON'T FORGET: OpenAI is cutting off Cursor's direct access to GPT models, with November 12 as the deadline as of now.
But you don't need to leave Cursor.
Just plug in a @ConcentrateAI API key, set our OpenAI-compatible base URL, and keep using OpenAI models in Cursor's Chat and Agent.
And never pay any fees.
https://t.co/zicmtzLsWF
This might be the coolest thing we've ever done at @concentrateai.
We flew out 18 founders to Austin for one week.
Here's what happened at The Builder House.
We've shipped a ridiculous amount over the last few weeks.
A few highlights:
- Dozens of new models and expanded provider coverage
- Completely redesigned Model Fortress with provider comparisons, live pricing, and advanced filters
- Dedicated model and provider pages with pricing, capabilities, and service tiers
- New routing controls, model allowlists, and configurable fallback chains
- Redesigned Playground for side-by-side model testing
- Expanded Flex and Priority inference options
- Better model aliases and provider pinning
- Improved Responses API compatibility, streaming, and tool calling
- More granular spend limits and API key permissions
- Dozens of reliability, caching, and performance fixes
🚢🚢🚢