
Evaluating Multimodal Large Language Models on Vertically Written Japanese Text
Description in other languages
Statistics
Statistics as of Nov 17, 2026
GitHub stars
3
±0 (Oct 13〜)
Last commit
May 2026
Created on GitHub
Aug 6, 2025
Listed
Mar 8, 2026
Stars over time
Show as a table
| Month | Stars |
|---|---|
| Dec 2025 | — |
| Jan 2026 | — |
| Feb 2026 | — |
| Mar 2026 | 1 |
| Apr 2026 | 1 |
| May 2026 | 2 |
| Jun 2026 | 3 |
| Jul 2026 | 3 |
| Aug 2026 | 3 |
| Sep 2026 | 3 |
| Oct 2026 | 3 |
| Nov 2026 | 3 |
Last snapshot of each month. Star counts of 1,000 or more are rounded to the nearest 100.
Popular in the same category
llm-leaderboardwandb95Project of llm evaluation to Japanese tasks
japanese-llm-rankingyuzu-ai50This repository supports YuzuAI's Rakuda leaderboard of Japanese LLMs, which is a Japanese-focused analogue of LMSYS' Vicuna eval.
llm-jp-eval-mmllm-jp44This tool automatically evaluates Japanese multi-modal large language models across multiple datasets.
japanese-llm-roleplay-benchmarkoshizo42This repository was created to evaluate the performance of character role-playing in Japanese LLM.
gpt4-autoevalnorthern-system-service17Script for automatically evaluating language model responses using GPT-4.
IgakuQA119naoto-iwase10Evaluating LLMs on the 119th Japanese Medical Licensing Examination
Resources in similar research fields
- vntl-leaderboardlmg-anon857A leaderboard that ranks large language models by their semantic accuracy in translating Japanese visual novels into English, using cosine similarity over 256 samples and reporting chrF scores for comparison with tools like Sugoi Translator, Google Translate, Naver Papago, and Alibaba Translate.HF datasetShared: Large language models・Evaluation & benchmarks
- llm-jp-4-vl-9bllm-jp4kLLM‑jp‑4‑VL 9B is a vision‑language model from LLM‑jp that builds on llm‑jp‑4‑8b‑thinking and SigLIP 2‑SO400m, trained on commercial‑friendly data, adds reasoning traces before answers, and outperforms its beta version on text and chart benchmarks such as CharXiv and HakushoBench.HF modelShared: Large language models・Vision & multimodal
IgakuQAjungokasai52Evaluating GPT-4 and ChatGPT on Japanese Medical Licensing ExaminationsShared: Large language models・Evaluation & benchmarks- PaddleOCR-VL-For-Mangajzhang5333kFine‑tuned from PaddleOCR‑VL, PaddleOCR‑VL‑For‑Manga achieves 70 % full‑sentence accuracy on Manga109‑s speech‑bubble crops—over triple the 27 % baseline—using a multi‑language dataset and includes training code and a developer guide.HF modelShared: Large language models・Vision & multimodal
- gemma-4-12B-it-qat-UD-japanese-imatrixdahara12kA 1/4‑size, CPU‑friendly Japanese‑optimized quantized Gemma 4 model (Apache 2.0) that runs entirely on-premise with optional developer support and robust benchmarks.HF modelShared: Large language models・Vision & multimodal
JaSPICEYuigaWada9JaSPICE: Automatic Evaluation Metric Using Predicate-Argument Structures for Image Captioning ModelsShared: Evaluation & benchmarks・Vision & multimodal