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

Find an embedding model that ranks relevant documents well for retrieval.

Worked example as of 2026-09-23, from the public data image.

Strongest retrieval evidence, cost no object.

Requirements

model.class = vectoriser: Retrieval embeddings require the model class that emits vectors.

Preferences

retrieval: Rank only by the retrieval capability estimate.

Applied template spec

spec_version

where

model.class = vectoriser

optimize

weights

retrieval: 1.0

Worked answer

Decision status: answered.

Separated on the evidence

qwen/qwen3-embedding-8b

probability bands: a model joins the leader when P(its score >= the leader's) >= 0.25 under the score posteriors; a weighted capability with an 80% interval wider than 2.8 is not enough evidence

Score and score interval use the objective's dimensionless normalised units. Cost per task is for the engine-selected offering and the applied spec's task size. When task size is omitted, the engine uses its documented default. P(best) is the engine probability. Unavailable means the engine did not return a value. Thin evidence is undetermined and is not a recommendation.

Best band

Model	Score	Score interval	P(best)	Cost per task, USD	Capability estimates and intervals
`qwen/qwen3-embedding-8b`	0.933479	[0.062415, 1.804542]	0.26171875	Unavailable	retrieval: 0.334151681; interval [-0.593146487, 1.26144985]; benchmarks 3; direct benchmarks 1

Rest band

Model	Score	Score interval	P(best)	Cost per task, USD	Capability estimates and intervals
`kingsoft/qzhou-embedding`	0.0	[-0.910881, 0.910881]	0.05078125	Unavailable	retrieval: -0.659590812; interval [-1.629276687, 0.310095063]; benchmarks 2; direct benchmarks 1

Thin evidence: undetermined

Model	Score	Score interval	P(best)	Cost per task, USD	Capability estimates and intervals
`microsoft/harrier-oss-v1-27b`	1.0	[-0.872477, 2.872477]	0.41015625	Unavailable	retrieval: 0.404967493; interval [-1.588393109, 2.398328095]; benchmarks 1; direct benchmarks 0
`tencent/kalm-embedding-gemma3-12b-2511`	0.600425	[-1.272051, 2.472902]	0.27734375	Unavailable	retrieval: -0.020402878; interval [-2.01376348, 1.972957723]; benchmarks 1; direct benchmarks 0

Engine warnings

not_separable

Decision dec_d3eb2f69b7ef1ea013efa71f on public snapshot snap_05a963d2477ac2fb.

What this template teaches

Model class is a Must, while measured retrieval quality is a Prefer.

The live board uses current data. Its answer can change as evidence changes.

Apply this template on Decide

Human lookup is free on Decide. CLI, API, and MCP access requires a key. Access and pricing.

Use-case examples · How we decide