FreshContextA context lab · Fife
Drawing
02 · Benchmarks
Sheet
2 of 3
Rev
A · 2026-10-01

Model · measured only← Models

k2-36b-extraction

3 recorded runs. This model has no written entry yet.

Scored runs
3
Run log entries
3
RecordIdentifier

k2-36b-extraction

Every result below is as recorded. Without a written entry there is no stated reason for the test and no verdict.

Recorded runs3 scored · 95% intervals
SuiteScore95% interval PassedRuntime · profileRecorded
nullschema54.2%42.7–65.2Passed 39/72llama.cppextraction2026-09-09
synthesis75.0%62.8–84.2Passed 45/60llama.cppextraction2026-09-09
worldview52.5%37.5–67.1Passed 21/40llama.cppextraction2026-09-09

Runs marked unenforced were measured through mlx_lm on a suite that requires constrained decoding. mlx_lm accepts the schema and ignores it, so those numbers do not measure what the suite is designed to measure. They are kept and labelled rather than deleted.

Every run log entry for this model (3)
FinishedSuiteRuntime · profileScoreCoverageGrader · dataset · config
synthesissynthesis/2not recordedextraction75.0%Coverage 60/60g 4156dd3c13361e8bd 267d1954e225c5b6c b92607640a462939
nullschemanullschema/1not recordedextraction54.2%Coverage 72/72g d7448e0f5fc3f9afd 3b91280bb40372efc 5e938f4fe9784ee9
worldviewworldview/2not recordedextraction52.5%Coverage 40/40g 9acfba1b704feb28d e97835de9e6fc227c 9cc504b6f261a5b3