Skip to content

Benchmark

NepNewsBench v2 draft entity precision (all 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

0.912
Mean
0.919
Median
0.028
Std dev
0.883
10th percentile
0.943
90th percentile
0.832
Min
0.8320.944
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) data0.874
Claude Haiku 4.5Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.832
Claude Opus 5Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.944
Claude Sonnet 4.6Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.909
Claude Sonnet 5Anthropic via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.933
Gemini 3 FlashGoogle AI Studio via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.915
Gemini 3.1 Pro PreviewGoogle AI Studio via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.903
Gemini 3.8 FlashGoogle AI Studio via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.923
Gemma 4 31B ITVenice via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.914
GLM 5.3Z.AI via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.933
GPT-5.6 LunaOpenAI via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.919
GPT-5.6 SolOpenAI via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.921
Grok 4.6xAI via openrouter, reasoning low· NepNewsBench v2 (draft candidate release) data0.934
Kimi K3Moonshot AI via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.943
Qwen 3.8 27BAlibaba via openrouter, reasoning off· NepNewsBench v2 (draft candidate release) data0.883
Qwen3.8 MaxAlibaba via openrouter, reasoning minimal· NepNewsBench v2 (draft candidate release) data0.917

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.