Component catalogue

156 records

2 of 156 match

Clear
Browse filters
resourceintake
Toloka

Human-in-the-loop training data, RL environments, and evaluation for AI agents and LLMs.

Function

Evaluate

score quality against a bench

Evaluate Browse Evaluate

Function

Classify

assign a category

Classify Browse Classify

Function

Generate

produce new text, code, or media

Generate Browse Generate

Function

Observe

watch a system

Observe Browse Observe

Interface

http

HTTP API

GitHub

No signal yet

Agent use

No signal yet

Reputation

Human and agent evidence, together.

Score factors

Toloka · 12

Score v2 uses observed signals only. Missing factors read No signal yet.

  • GitHub healthNo signal yet
  • Agent useNo signal yet
  • Human notes1.2 pts
  • ComposabilityNo signal yet
  • Checklist6.7 pts
  • Evidence quality4.6 pts
  • BenchmarksNo signal yet
Full Score explanation
Observed signals only
modelintake
TypeSafe Jev

Returns typed judgments and probabilities from software state and narrowly defined questions.

Function

Classify

assign a category

Classify Browse Classify

Function

Evaluate

score quality against a bench

Evaluate Browse Evaluate

Function

Reason

plan or chain steps

Reason Browse Reason

Interface

http

HTTP API

Open spec

GitHub

No signal yet

Agent use

No signal yet

Reputation

Human and agent evidence, together.

Score factors

TypeSafe Jev · 7

Score v2 uses observed signals only. Missing factors read No signal yet.

  • GitHub healthNo signal yet
  • Agent useNo signal yet
  • Human notesNo signal yet
  • ComposabilityNo signal yet
  • Checklist3.3 pts
  • Evidence quality3.7 pts
  • BenchmarksNo signal yet
Full Score explanation
Observed signals only

Evidence

Leyline ranks components. It does not proxy inference or resell tokens.

Human evidence

Observed stars, notes, maintainer trust, and public sources. Missing signals read No signal yet.

Agent evidence

Observed installs, success, resolve wins, and benches. Unobserved usage is never guessed or zeroed.