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

Model · measured only← Models

k2-4b-bf16-precision

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

Scored runs
5
Run log entries
5
RecordIdentifier

k2-4b-bf16-precision

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

Recorded runs5 scored · 95% intervals
SuiteScore95% interval PassedRuntime · profileRecorded
nullschema43.1%32.3–54.6Passed 31/72llama.cppextraction2026-09-10
synthesis56.7%44.1–68.4Passed 34/60llama.cppextraction2026-09-10
toolcall100.0%94.0–100.0Passed 60/60llama.cppextraction2026-09-10
toolscale100.0%91.2–100.0Passed 40/40llama.cppextraction2026-09-10
worldview50.0%35.2–64.8Passed 20/40llama.cppextraction2026-09-10

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 (5)
FinishedSuiteRuntime · profileScoreCoverageGrader · dataset · config
synthesissynthesis/2not recordedextraction56.7%Coverage 60/60g 4156dd3c13361e8bd 267d1954e225c5b6c 0faa5511a9576d51
nullschemanullschema/1not recordedextraction43.1%Coverage 72/72g d7448e0f5fc3f9afd 3b91280bb40372efc 776fc6d33aa8cc34
worldviewworldview/2not recordedextraction50.0%Coverage 40/40g 9acfba1b704feb28d e97835de9e6fc227c 770c2ce3d2f16e96
toolscaletoolscale/1not recordedextraction100.0%Coverage 40/40g a2bc3751abfa28e0d b24abb108c409178c e21f6ac8887f7f5d
toolcalltoolcall/1not recordedextraction100.0%Coverage 60/60g 804e8fc9457649d8d 10af88c170513155c 2c2e5971d9f6382f