Skip to content

Benchmark

BABILong qa2 512k

Long context benchmarks measure how well models use information in long inputs. Scores are percentages. Higher is better.

Models and systems measured
3
Scores recorded
3
Sources
1
IQ weight
No weight published

Score distribution

43.3%
Mean
25%
Median
40.1%
Std dev
6%
10th percentile
99%
90th percentile
6%
Min
6%99%
The band covers the middle 80% of scores. The line marks the median.

Results

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
~ ARMT (137M) fine-tuneqa2, 512k· BABILong v0 data docs99%
~ RMT (137M) fine-tuneqa2, 512k· BABILong v0 data docs25%
Llama3-ChatQA-1.5-8B + RAGqa2, 512k· BABILong v0 data docs6%

Data credits: RMT-team, Kuratov et al. Official v0 numeric CSV; results licence not separately stated.