Hands-on with Tencent Hy-MT2 1.8B Q4_K_M (1.08GB) on M1 Max 64GB via llama-server. JSON, SRT, HTML, glossary, and minority-language prompts with full input-output pairs. The 1.25bit 440MB build does not load on stock llama.cpp 8990, and 30B-A3B (hy_v3) is not in the Mac route yet.
MyMemory is a free translation API you can fetch straight from client-side JS on a static site. Endpoint spec, language pairs, the 500-byte request limit, daily quotas with and without an email, and a working fetch example with error handling.
