Rank,Model,Type,BLEU,chrF++,Our Model
1,Google Translate,Commercial Service,9.27,46.44,FALSE
2,Claude Sonnet 4,Anthropic,8.07,43.54,FALSE
3,Gemini 2.0 Flash 001,Google,7.94,43.38,FALSE
4,Gemini 2.5 Pro,Google,7.46,44.74,FALSE
5,Ganda Gemma 1B (Our Model),Specialized Fine-tuned,6.99,40.32,TRUE
6,Gemini 2.5 Flash,Google,6.28,40.51,FALSE
7,Chatgpt 4o Latest,OpenAI,6.19,40,FALSE
8,Gpt Oss 120B,OpenAI,5.16,34.7,FALSE
9,Gpt 5 Mini,OpenAI,5.14,36.55,FALSE
10,Gpt 5 Nano,OpenAI,4.93,32.86,FALSE
11,Llama 4 Maverick,Meta,4.52,33.75,FALSE
12,Gemma 3 27B,Google,3.65,31.37,FALSE
13,Llama 4 Scout,Meta,3.59,27.63,FALSE
14,Gpt Oss 20B,OpenAI,3.39,27.98,FALSE
15,Gemma 3 4B,Google,1.1,20.05,FALSE
16,Gemma 3N E4B,Google,0.84,17.6,FALSE
17,Gemma 3 1B (vLLM Baseline),General Model,0.51,9.79,FALSE