Skip to content
Longterm Wiki
benchmark-result

DeepSeek Models on GSM8K: 89.3

Child of GSM8K

Metadata

Source Tablebenchmark_results
Source IDxIGxFaG3fF
ParentGSM8K
Children
CreatedApr 24, 2026, 6:53 PM
UpdatedApr 24, 2026, 6:53 PM
SyncedApr 24, 2026, 6:53 PM

Record Data

idxIGxFaG3fF
benchmarkIdfjjBrOI3p2
modelIdDeepSeek Models(ai-model)
score89.3
unitpercent
date2024-12-01
sourceUrl
notesDeepSeek-V3 GSM8K accuracy, 8-shot evaluation

Source Check Verdicts

confirmed99% confidence

Last checked: 4/24/2026

Inline sourcing: confirmed

Debug info

Thing ID: xIGxFaG3fF

Source Table: benchmark_results

Source ID: xIGxFaG3fF

Parent Thing ID: fjjBrOI3p2