Canarim 7B
Brazilian Portuguese 7B LLM with extended training on Portuguese web text.
About
Canarim 7B (dominguesm/canarim-7b) is a Brazilian Portuguese decoder-only language model developed by Miguel Domingues, released in 2023. It extends training with 16 billion tokens from the Portuguese Common Crawl, targeting fluent Portuguese generation. The release date below is era-level (2023) since only year granularity is verifiable.
The project also produced Canarim-7B-VestibulAide, a variant tuned for Brazilian university entrance exam questions, and the Canarim Instruct PT-BR dataset of over 300k Portuguese instructions used by the community for fine-tuning.
Canarim is part of the first wave of Brazilian community LLMs documented in the Brazilian LLM literature, alongside Cabrita, Bode, and Sabiá.
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 — 6/10
Scored 6/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 111; Trending score 58/100 from researcher assessment.