Trade Your Capacity Like an Asset.

The single platform where AI model providers place, reprice and cancel orders on models and activity types, serve live interactive demand, run algorithmic yield, and settle every trade with an auditable record. Micro-batching is there when you want flexible work paced onto the fleet.

Why supply on Keld

Three Reasons to List Your Fleet.

Monetize Idle Capacity

Put spare GPU throughput to work. Idle blocks clear into live enterprise demand and become billable revenue, a new channel alongside your existing ones.

Get Discovered

Models delivering near-frontier quality at a fraction of the cost get found and used, not overlooked. Specialist and near-frontier fleets are matched on value, not brand.

Serve Any Kind of Demand

Most volume arriving on Keld is live interactive work: coding, chat, text creation. Take it as it comes, and when you'd rather smooth it, our micro-batching layer paces flexible work onto your fleet, so you offer batch inference without building it.

Orders

Work the Book,
in Real Time.

Post asks on a named model or on an activity type, adjust depth and reprice in seconds as fleet utilization shifts, so idle GPU blocks clear without eroding your premium direct pricing.

  • Place, cancel and reprice resting asks against live book depth
  • Two kinds of book: an individual model book, or a Keld-curated activity-type book
  • SDK, graphical UI and an autonomous trading-agent layer
  • Algorithmic yield management that works the book for you
  • No lock-in: list on Keld alongside any other channel
Order Book● Live
Model tierDemand$/1MRev/day
Llama 3.3 70B
$0.52$8,100
DeepSeek V3
$0.28$4,300
Qwen2.5 72B
$0.40$3,200
Custom 8B
$0.18low demand
Micro-batching · optional

Smooth the Stream
into Your Fleet.

Interactive work streams straight through to your endpoints. For the flexible share, a micro-batch controller run by Keld on your behalf paces the matched stream, absorbing demand spikes and turning idle headroom into yield, so you can offer batch inference without building the infrastructure for it. Use it where it helps; it is one way to work with Keld, not the only one.

  • Interactive activities pass through unbatched, at market price
  • Flexible activities paced to live per-node utilization
  • Queueing, batching, pacing, retries and backpressure, all handled
  • Higher effective yield, zero dropped work, no batching infrastructure to maintain
Fleet pacing● 92% utilization
Raw bursty
stream
Paced
micro-batches
Fleet at steady
utilization
node a100-01
90%
node a100-02
94%
node h100-01
88%
queue · micro-batched
paced
Settlement

Automated Settlement,
Full Audit Trail.

Keld handles clearing and settlement end-to-end, so you focus on running infrastructure.

  • Every order independently auditable, end-to-end
  • Per-order records: model, volume, price, timestamp, buyer
  • Settlement summaries exportable to your finance stack
  • Dispute resolution backed by immutable order records
Settlement Log● Settled
OrderModelAmountStatus
#4821Llama 3.3 70B$1,240settled
#4820DeepSeek V3$380settled
#4819Qwen2.5 72B$2,100settled
#4818Custom 8B$94clearing
Discovery

Get Found on
Value, Not Brand.

Enterprises building apps and agents send live work to Keld, and your capacity sits in front of them. Near-frontier and specialist models are surfaced by capability and price, so the fleets that deliver get used, not overlooked. Execution results feed back into how each book is ranked, so measured quality, not marketing, wins the flow.

  • Activities match on price, measured quality and fit for the activity type, never on brand
  • Near-frontier quality at a fraction of the cost gets found
  • Buyers discover your fleet by capability, not marketing spend
  • Named-model guarantees enforced end-to-end
Best match for this job● Surfaced on value
Model ranked by valueQuality$/1MRank
Nimbus-8B · specialist, little-known
$0.09★ #1
Big-Brand-70B · well-known
$0.60#4
Frontier-400B · famous
$2.40#9 · overkill

New Revenue Stream

Sell idle compute into live demand, a channel alongside your existing ones.

One Platform, Both Planes

The financial plane (orders, yield, settlement) unified with the data plane delivering matched work — streamed live, or micro-batched when you prefer.

No Lock-in

List on Keld alongside any other channel. Named-model guarantees enforced end-to-end.

List Your Fleet on Keld.

Apply to supply, connect your endpoints, and post your first resting asks on a model or an activity type. Keld delivers matched work to your fleet — streamed live or micro-batched, your call — and settles every trade automatically. No lock-in.