Hy-MT2-1.8B is a large language model released by Tencent, released under the apache-2.0 license, with a claimed knowledge cutoff of 2026-05-21. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.
Description
Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent's Hunyuan team, open-sourced May 21, 2026 alongside 7B and 30B-A3B siblings. It supports translation across 33 languages plus Chinese dialect/minority-language pairs, can be quantized to as little as 440MB, and is reported to outperform several mainstream commercial translation APIs despite its size.
Metadata
Hugging Face stats
28,477 downloads on Hugging Face · 1,222 likes · pipeline: translation · library: transformers
Weight Hashes
Weights file
SHA-25629e9117a44c79f81857613601968ff482d8a23c2d6736a1710bba9e5ca4762e5
SHA3-256357bb6cdff5a84d4360e01e5c2e80f77e1b614e0e742c66721d3315a4c734f91
BLAKE39c1b7e4a03f30589da48bb36c42b77048be3018159bed9bf1935b50c2df256e4
Config file
SHA-256da40c514cc74a5748a2e591b1b95fca4b7e94de05349abe4ea4164a82641de1a
SHA3-256e61ffee16543dcbf255174c7ba370cb9aed67c0581a27f1f0dfb5537f4a4a416
BLAKE39ee40996ed0352bc439b36b012d22005a2646a3446f7b74d1e20b67c676892a3
Tokenizer file
SHA-256b475bbef1b0b2fd57dcb865332b546475bd1ede2deb3bb91bafd0c047a8a530a
SHA3-2560318ae5d3646817d770b9c20dbd0732b0bba9a30750a28518b8fbe307e0e3402
BLAKE373fefe7983ddc07b18b60452afdc19857bf3fde18d9aca7a994a6f1c75a7828a
Verify it yourself
curl -L -o "model.safetensors" "https://huggingface.co/tencent/Hy-MT2-1.8B/resolve/9a341cd1b679d3efd23b46e847b01745a71ed792/model.safetensors" sha256sum "model.safetensors"
Expected SHA-256: 29e9117a44c79f81857613601968ff482d8a23c2d6736a1710bba9e5ca4762e5