PhoGPT-7B5
Vietnam's 7.5B open Vietnamese LLM, pretrained from scratch on 41GB of text.
About
PhoGPT-7B5 is a 7.5-billion-parameter Vietnamese language model from VinAI (Vingroup), unveiled at AI Day 2023 on December 5. Unlike adapted multilingual models, it was pretrained from scratch on a 41GB Vietnamese corpus with a Vietnamese-specific byte-level BPE tokenizer, using flash attention and ALiBi for context extrapolation. An instruction-tuned variant, PhoGPT-7B5-Instruct, was fine-tuned on 150K Vietnamese prompt-response pairs.
VinAI reported performance near GPT-3.5-turbo on Vietnamese evaluations and released it open-source for commercial use. The team later recommended the smaller PhoGPT-4B series for better efficiency. Technical report: arXiv 2311.02945.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 119; Trending score 66/100 from researcher assessment.