ModelSpec is an analysis of alternatives for AI models: it decides which models fit a job from your requirements, sourced benchmarks and real cost, and shows its work.
Retrieval embeddings, fastest
Find the hosted embedding model with the fastest offering, among models within 1.0 of the strongest retrieval estimate.
Worked example as of 2026-09-23, from the public data image.
Fastest hosted embedding among the strong; only models within 1.0 of the strongest retrieval estimate compete.
Requirements
model.class = vectoriser: Retrieval embeddings require the model class that emits vectors.retrieval >= best(1.0): Keep only models whose retrieval estimate is within 1.0 of the strongest eligible model's; speed decides among them. A gap of 1.0 can be measurable but is not decisive.
Preferences
offering.speed.throughput: Among the models the floor keeps, prefer the offering that produces output tokens fastest.
Applied template spec
spec_version- 1
where- model.class = vectoriser
- retrieval >= best(1.0)
optimizeweightsoffering.speed.throughput- 1.0
Worked answer
Decision status: no_feasible.
No recommendation: the public image has no answer with enough evidence.
Unknown requirements
Some candidates may qualify. Missing facts:
offering.speed.throughput
Missing evidence and possible changes
- no model that meets the requirements has a value for offering.speed.throughput
Decision dec_40efecc1f2c5412980af2716 on public snapshot snap_05a963d2477ac2fb.
What this template teaches
Strength is a floor relative to the strongest eligible model, so speed can never buy it back. Speed is a Prefer, not a gate. An offering without a measured speed may qualify but is not ranked, so this tier answers only once speeds are published.
The live board uses current data. Its answer can change as evidence changes.
Human lookup is free on Decide. CLI, API, and MCP access requires a key. Access and pricing.