Compare runs
Pick two or more runs to compare. Qwen3.8-27B-Fast on an unnamed computer is the first; add another below.
| Run | Qwen3.8-27B-FastNo computer named · llama-swapTake out Qwen3.8-27B-Fast |
|---|---|
| Score | 2,431Silver |
| Tasks passed | 44 of 44 |
| Points | 64 of 64 |
| Speed | 24.3 tok/s |
| Tokens | 54,114 |
| Summarizing | 5/5 |
| One-sentence summary | Passed24 tok/s |
| Meeting notes in three bullets | Passed24.5 tok/s |
| Review sentiment and complaint | Passed23.5 tok/s |
| Faithful summary of a study | Passed24.5 tok/s |
| Action items from an email thread | Passed24.4 tok/s |
| Extracting | 5/5 |
| Fields from an email as JSON | Passed24.8 tok/s |
| Every email address, once | Passed24.6 tok/s |
| A list as CSV | Passed24.5 tok/s |
| The real deadline | Passed25.4 tok/s |
| Warnings and errors from a log | Passed25.2 tok/s |
| Following instructions | 5/5 |
| Exactly one word | Passed25.1 tok/s |
| Three lowercase lines | Passed23.6 tok/s |
| Exact JSON shape | Passed25.1 tok/s |
| Sort a list | Passed24.5 tok/s |
| A name with three rules | Passed24.1 tok/s |
| Reasoning | 5/5 |
| Unit price | Passed24.8 tok/s |
| Days of the week | Passed24.6 tok/s |
| Who owns the cat | Passed24.4 tok/s |
| Arrival time | Passed25.2 tok/s |
| Filling a tank | Passed24.4 tok/s |
| Reading code | 4/4 |
| What does it print | Passed25.2 tok/s |
| Trace a loop | Passed25.2 tok/s |
| Find the bug | Passed24.2 tok/s |
| Run a query in your head | Passed25.2 tok/s |
| Using tools | 5/5 |
| Call one tool | Passed24.9 tok/s |
| Don't call a tool when none is needed | Passed26 tok/s |
| Two calls for two answers | Passed25.6 tok/s |
| Chain two tools | Passed25.1 tok/s |
| Handle a failed call honestly | Passed24.8 tok/s |
| Building pages | 5/5 |
| A small website | Passed24.6 tok/s |
| A pricing table with a toggle | Passed25.1 tok/s |
| An accessible sign-up form | Passed24.8 tok/s |
| A to-do app that remembers | Passed25.1 tok/s |
| A bar chart in SVG | Passed24.7 tok/s |
| Building games | 2/2 |
| Build 2048 | Passed24.4 tok/s |
| Build Snake | Passed23.7 tok/s |
| Writing code | 3/3 |
| Merge intervals | Passed24.2 tok/s |
| An LRU cache | Passed23 tok/s |
| Evaluate arithmetic | Passed23.7 tok/s |
| Fixing code | 2/2 |
| Fix the bookings module | Passed24.2 tok/s |
| Fix the CSV reader | Passed24.1 tok/s |
| Planning and data | 3/3 |
| A query over hidden data | Passed23.7 tok/s |
| Schedule five meetings | Passed23.6 tok/s |
| A replacement in ten calls or fewer | Passed23.9 tok/s |
Scrolls sideways to show every run.
A check marks the best of each row. tok/s is tokens a second: how fast the model writes. How the score works.