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

Model · measured only← Models

reap-60b-mlx-4bit-uniform

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

Scored runs
6
Run log entries
6
RecordIdentifier

reap-60b-mlx-4bit-uniform

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

Recorded runs6 scored · 95% intervals
SuiteScore95% interval PassedRuntime · profileRecorded
nullschemaunenforced84.7%74.7–91.2Passed 61/72mlxextraction2026-09-11
synthesis85.0%73.9–91.9Passed 51/60mlxextraction2026-09-11
toolcall28.3%18.5–40.8Passed 17/60mlxextraction-t072026-09-11
toolcall25.0%15.8–37.2Passed 15/60mlxextraction2026-09-11
toolscale95.0%83.5–98.6Passed 38/40mlxextraction2026-09-11
worldview47.5%32.9–62.5Passed 19/40mlxextraction2026-09-11

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 (6)
FinishedSuiteRuntime · profileScoreCoverageGrader · dataset · config
toolcalltoolcall/1mlxextraction-t0728.3%Coverage 60/60g 804e8fc9457649d8d 10af88c170513155c d696a3a98ce65e4f
synthesissynthesis/2mlxextraction85.0%Coverage 60/60g 4156dd3c13361e8bd 267d1954e225c5b6c 5f382a39bf24731a
nullschemanullschema/1mlxextraction84.7%Coverage 72/72g d7448e0f5fc3f9afd 3b91280bb40372efc 52349074a17dc8f5
worldviewworldview/2mlxextraction47.5%Coverage 40/40g 9acfba1b704feb28d e97835de9e6fc227c 76804a68d58c5eb6
toolscaletoolscale/1mlxextraction95.0%Coverage 40/40g a2bc3751abfa28e0d b24abb108c409178c a58faee14a0f3ed6
toolcalltoolcall/1mlxextraction25.0%Coverage 60/60g 804e8fc9457649d8d 10af88c170513155c 9fc8dc4dd468ed18