Skip to content

Benchmark

NepNewsBench v2 draft entity recall (Nepali-source questions)

Multilingual benchmarks compare model performance across languages. Higher scores are better.

Models measured
16
Scores recorded
16
Sources
1
IQ weight
No weight published

Score distribution

0.583
Mean
0.565
Median
0.09
Std dev
0.521
10th percentile
0.734
90th percentile
0.382
Min
0.3820.759
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.

ModelScore
Claude Fable 5.1Anthropic via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.759
Claude Haiku 4.5Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.382
Claude Opus 5Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.734
Claude Sonnet 4.6Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.575
Claude Sonnet 5Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.639
Gemini 3 FlashGoogle AI Studio via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.565
Gemini 3.1 Pro PreviewGoogle AI Studio via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.559
Gemini 3.8 FlashGoogle AI Studio via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.564
Gemma 4 31B ITVenice via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.521
GLM 5.3Z.AI via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.611
GPT-5.6 LunaOpenAI via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.565
GPT-5.6 SolOpenAI via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.561
Grok 4.6xAI via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.629
Kimi K3Moonshot AI via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.651
Qwen 3.8 27BAlibaba via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.443
Qwen3.8 MaxAlibaba via openrouter, reasoning minimal· NepNewsBench v2 (draft candidate release) data0.563

Data credits: Outback Yak Research, NepNewsBench v2 research draft. Public numerical aggregates from a candidate preview that the publisher says is not evidence of publication; no release licence assigned and none asserted here.