Brett Harrison

Liquid Inference now has a Discover tab with more info on which models received the most routing traffic. Inference providers have more info to optimize their offering.

Going on MTS at 4:15ET/1:15PT to discuss the launch of Liquid Inference and price discovery mechanisms for inference tokens. Looking forward!

As a BD normie, I've been using the "Compare" feature on Liquid Inference most and it's pretty insightful. Way too much friction to get these insights without it. I'd have to create multiple accounts, prompt multiple tabs, copy/paste etc.

With Liquidinference.ai, I see them running side by side in real time, with the output, cost, speed all there.

This is so much more meaningful than any performance test or benchmarks because I'm developing my bespoke benchmarks that can't be gamed.

It's not just for power users. Most businesses can save by efficiently routing inference for things like a social post announcing an event. We're gonna be at Rosenblatt's Fintech Summit btw, post incoming.

We're live and it's a huge day - here's what's happening:

- Nick Bostrom is joining MTS today at 930am PT for a full hour of conversation
- Isomorphic Labs in talks to raise at a $40B+ valuation
- Google launches the Gemini agent, a universal AI agent for work
- Arena AI raises $200m and releases a new Alignment Index
- TSMC posts record Q3 revenue, up 50% on AI chip demand

Followed by other great interviews throughout the day:

- @real_ioannis Co-founder and CTO at Reflection AI talking about Beam, Reflection's latest 501B open-weight model

- @nathanbenaich Founder and General Partner at Air Street Capital talking about the State of AI Report 2026, out today

- @natolambert Co-founder and Executive Director at Trillium Labs talking about his new nonprofit for open science of frontier AI

- @Brendan_McCord Founder at Cosmos Ventures talking about its $77.6M Fund I for philosopher-builders

- @sytaylor Founder at Brainfood talking about his essay "Fintech is Dead"

- @ml_angelopoulos Co-founder and CEO at Arena talking about their $200m raise at a $3.1b valuation, and the release of their Alignment Index

- @nautsimon_ Founding Member of Technical Staff, @a_ramabadran Member of Technical Staff and @tobiges Co-Founder and Head of Reinforcement Learning at Axiom Math talking about DrivingBench, where frontier AI models drive a real car

- @BrettHarrison Founder and CEO at Architect talking about Liquid Inference, a live exchange where providers compete on price for every AI prompt

- @PrinehaN Professor at UCLA and Operating Partner at DCVC talking about DARPA moving four more teams into the final stage of its Quantum Benchmarking Initiative

- @ayushswrites Founder and CEO at Warp talking about Warp 2.0 and AI agents for HR and payroll

We're live and it's a huge day:

Nick Bostrom is joining MTS today at 9:30am PT for a full hour of conversation, followed by

@real_ioannis - Co-Founder and CTO at Reflection AI talking about Beam, Reflection's latest 501B open-weight model

@nathanbenaich - Founder and General Partner at Air Street Capital talking about the State of AI Report 2026, out today

@natolambert - Co-Founder and Executive Director at Trillium Labs talking about his new nonprofit

@Brendan_McCord - Founder at Cosmos Ventures talking about its $77.6M Fund I for philosopher-builders

@sytaylor - Founder at Brainfood talking about his essay "Fintech is Dead"

@ml_angelopoulos - Co-founder and CEO at Arena talking about their $200m raise at a $3.1b valuation, and the release of their Alignment Index

@nautsimon_ - Founding Member of Technical Staff,
@a_ramabadran, Member of Technical Staff and
@tobiges, Member of Technical Staff at Axiom Math talking about DrivingBench, where frontier AI models drive a real car

@BrettHarrison - Founder and CEO at Architect talking about Liquid Inference

@PrinehaN - Professor at UCLA and Operating Partner at DCVC talking about DARPA moving four more teams into the final stage of its Quantum Benchmarking Initiative

@ayushswrites - Founder and CEO at Warp talking about Warp 2.0 and AI agents for HR and payroll

It’s been one day since the launch of Liquid Inference and we’ve doubled our provider network. We’re cheaper than OpenRouter for Kimi K3, DeepSeek V4.1 Flash, Qwen3.8 27B, and ~30 more models. Every new provider that joins increases price competition.

In July we wrote that GPU-hours were the first compute surface to financialize and inference tokens would come next.

@Architect_Fi shipped the second leg way ahead of schedule, and 2/6 launch providers are crypto networks.

In our report we argued the realistic onchain path was plugging decentralized supply into Architect's hub as capacity.

Congrats @BrettHarrison and team

RFQ for tokens!

The response to LI has been overwhelmingly positive. I had a first call with a new inference provider at 5pm ET and they're live and quoting on the platform. Wild.

Proud to be a day-1 provider on @Architect_Fi. Inference need real markets and we're bringing capacity to help build them. Congrats @BrettHarrison and team.

Say it once more for the folks in the back: compute is a commodity.

Nobody buys oil or electricity on a monthly subscription. Free yourself from the SaaS paradigm of bundled services and high margins. There's a better way, and it's Liquid Inference. Check it out, first 500 get $20 in free tokens liquidinference.ai.

Leverage the power of market competition for the best possible pricing. Purchase only what's needed. Substitute models and switch suppliers to control costs.

Buyers pay the lowest quote per prompt across any subset of over 700 models. Providers can sell spare GPU capacity and create a new sales channel.

It was good to work with a team like @Architect_Fi (@Ishanee and @BrettHarrison in particular) who move at our kind of speed!

Looking forward to serving a ton of new customers.

Yano asks whether adding a futures market around compute could create risk of someone manipulating the benchmark to trigger a selloff.

@BrettHarrison argues opacity is the bigger danger. The answer is transparent pricing and an index that’s hard to manipulate.

Excited to be one of the launch partners! We’re building toward a future where compute is easier to access, manage, and trade.

Congrats on the launch. Looking forward to what we’ll build together!

Looking forward to support @BrettHarrison with @engyai to make inference markets more efficient.

Inference need proper markets and im certain Brett is one of the most qualified persons out there to figure this out.

Inference markets have a lot of room for innovation. Excited to contribute through @zroai_ to this new product, pushing inference routing toward economic efficiency.
Congrats @BrettHarrison and Architect

AkashML is the launch partner for @Architect_Fi’s Liquid Inference.

Every prompt is auctioned, and providers compete on price. The request routes to the lowest offer that meets the buyer’s rules.

OpenAI and Anthropic SDKs are compatible with one API key, where the price is locked before the first token.

Decentralized GPUs are now in a live inference market.

Pearl @prlnet models are now live through Architect’s new inference marketplace (serving DeepseekV41-Flash)

Introducing Liquid Inference by Architect, a new LLM model router where inference providers compete dynamically to offer lowest-cost tokens. Providers onboard in minutes, not weeks. Buyers have full price visibility and receive 20% of referred fees in free inference. More info:

Liquid Inference can be used in place of any OpenAI or Anthropic API-compatible backend. Behind it, our team has leveraged our experience trading and building financial exchanges to create real-time two-sided price discovery for inference. Benefits for buyers and providers:

INFERENCE BUYERS

• Fully compatible with agentic coding tools: Claude Code, Codex, OpenCode, Cursor, Pi, Cline, and many more. Full multi-modal support.
• Hundreds of open- and closed-weight models. Create routing rule presets or use our Auto routing algorithms.
• Providers compete dynamically on price for every prompt, offering the lowest marginal cost.
• Sign up for free with email, first 500 users get $20 of free inference.
• Receive 20% of referred fees as free inference, 10% for second-level referrals.

INFERENCE PROVIDERS

• Onboard through the Liquid Inference app. We verify new providers in minutes, not weeks.
• All prompts are OpenAI API standard. Simple REST/Websocket API to register models and quotes.
• Update quotes dynamically based on your own costs. Monetize GPU capacity only when you want.
• Standardize payouts via Stripe, full itemized records of all jobs performed.

Today we’re launching Liquid Inference, an OpenAI/Anthropic-compatible router where inference providers compete on price for every prompt. Buyers route across 700+ open- and closed-weight models at the lowest available marginal cost, providers onboard instantly to start serving models.