Hacker News new | ask | show | jobs
Dynamo 8B: multilingual LLM for global enterprises (dynamofl.com)
13 points by _eric_z_lin 897 days ago
1 comments

Dynamo 8B outperforms Mistral 7B, Llama2 13B, Bloom 7B, and PolyLM 13B on most multilingual benchmarks we tested (i.e. PAWS and XCOPA).

Dynamo 8B has not been instruction fine-tuned and has not undergone alignment using techniques like reinforcement learning from human feedback. The intention behind crafting this model is to provide the research community with a model to explore vital multilingual capabilities that enable widespread use of LLMs globally.