Pre-built tokie tokenizer for mistralai/Mixtral-8x7B-v0.1.
pip install tokie
import tokie
tokenizer = tokie.Tokenizer.from_pretrained("tokiers/Mixtral-8x7B-v0.1")
encoding = tokenizer.encode("Hello, world!")
print(encoding.ids)
print(encoding.attention_mask)
[dependencies]
tokie = { version = "0.0.4", features = ["hf"] }
use tokie::Tokenizer;
let tokenizer = Tokenizer::from_pretrained("tokiers/Mixtral-8x7B-v0.1").unwrap();
let encoding = tokenizer.encode("Hello, world!", true);
println!("{:?}", encoding.ids);
tokenizer.tkz β tokie binary format (~10x smaller, loads in ~5ms)tokenizer.json β original HuggingFace tokenizer50x faster tokenization, 10x smaller model files, 100% accurate.
tokie is a drop-in replacement for HuggingFace tokenizers, built in Rust. See GitHub for benchmarks and documentation.
MIT OR Apache-2.0 (tokie library). Original model files retain their original license from mistralai/Mixtral-8x7B-v0.1.
8 commits
Pre-built tokie tokenizer for mistralai/Mixtral-8x7B-v0.1.
pip install tokie
import tokie
tokenizer = tokie.Tokenizer.from_pretrained("tokiers/Mixtral-8x7B-v0.1")
encoding = tokenizer.encode("Hello, world!")
print(encoding.ids)
print(encoding.attention_mask)
[dependencies]
tokie = { version = "0.0.4", features = ["hf"] }
use tokie::Tokenizer;
let tokenizer = Tokenizer::from_pretrained("tokiers/Mixtral-8x7B-v0.1").unwrap();
let encoding = tokenizer.encode("Hello, world!", true);
println!("{:?}", encoding.ids);
tokenizer.tkz β tokie binary format (~10x smaller, loads in ~5ms)tokenizer.json β original HuggingFace tokenizer50x faster tokenization, 10x smaller model files, 100% accurate.
tokie is a drop-in replacement for HuggingFace tokenizers, built in Rust. See GitHub for benchmarks and documentation.
MIT OR Apache-2.0 (tokie library). Original model files retain their original license from mistralai/Mixtral-8x7B-v0.1.
8 commits