abhinand/gemma-2b-it-tamil-v0.1-alpha
TEXT GENERATIONConcurrency Cost:1Model Size:2.6BQuant:BF16Ctx Length:8kPublished:Feb 25, 2024License:gemma-terms-of-useArchitecture:Transformer0.0K Warm

The abhinand/gemma-2b-it-tamil-v0.1-alpha is a 2.6 billion parameter instruction-tuned model developed by Abhinand Balachandran, based on Google's Gemma 2B architecture. This experimental model is specifically fine-tuned for bilingual English and Tamil language tasks, demonstrating adaptation for Tamil without expanding its vocabulary. It shows promising performance, outperforming Google's Gemma 2B base and instruct models, as well as mlabonne/Gemmalpaca-2B, on various benchmarks.

Loading preview...