resyn

Compare runs

Pick two or more runs to compare. Qwen3-4B-Instruct-Abliterated-Transform on an unnamed computer is the first; add another below.

Runs of the same model, or on the same computer, the newest first.

RunQwen3-4B-Instruct-Abliterated-TransformNo computer named · Ollama v262Take out
Score—No score
Tasks passed22 of 44
Points22 of 64
Speed87.2 tok/s
Tokens10,196
Summarizing4/5
One-sentence summary83 tok/s
Meeting notes in three bullets88.5 tok/s
Review sentiment and complaint86.8 tok/s
Faithful summary of a study84.9 tok/s
Action items from an email thread88.2 tok/s
Extracting4/5
Fields from an email as JSON88.5 tok/s
Every email address, once
A list as CSV86.4 tok/s
The real deadline80.7 tok/s
Warnings and errors from a log87.2 tok/s
Following instructions4/5
Exactly one word86.2 tok/s
Three lowercase lines86.2 tok/s
Exact JSON shape86.9 tok/s
Sort a list87.6 tok/s
A name with three rules83.1 tok/s
Reasoning0/5
Unit price
Days of the week
Who owns the cat
Arrival time
Filling a tank
Reading code1/4
What does it print
Trace a loop
Find the bug87.2 tok/s
Run a query in your head
Using tools5/5
Call one tool86.2 tok/s
Don't call a tool when none is needed84.9 tok/s
Two calls for two answers85.5 tok/s
Chain two tools86 tok/s
Handle a failed call honestly86.2 tok/s
Building pages4/5
A small website87.6 tok/s
A pricing table with a toggle86.9 tok/s
An accessible sign-up form87.6 tok/s
A to-do app that remembers87.8 tok/s
A bar chart in SVG
Building games0/2
Build 2048
Build Snake
Writing code0/3
Merge intervals
An LRU cache
Evaluate arithmetic
Fixing code0/2
Fix the bookings module
Fix the CSV reader
Planning and data0/3
A query over hidden data
Schedule five meetings
A replacement in ten calls or fewer

A check marks the best of each row. tok/s is tokens a second: how fast the model writes. How the score works.