Tencent's Hy-MT2 Hits GitHub with Training Code and IFMTBench Benchmark
aigclink · x · 2026-10-09
Tencent Hunyuan's Hy-MT2 translation models are now on GitHub (Tencent-Hunyuan/Hy-MT2), already at 1.2k stars with 110 forks. The repo includes IFMTBench (an instruction-following translation benchmark), a train directory, bilingual READMEs, and links to the technical report.
The family offers 1.8B, 7B, and 30B-A3B (MoE) sizes covering 33 languages; with AngelSlim 1.25-bit quantization the 1.8B model needs only 440MB and runs 1.5x faster. Multi-dimensional evaluations show the 7B and 30B-A3B outperform open-source models such as DeepSeek.
Related event: Tencent Open-Sources Hy-MT2 Multilingual Translation Models(2 posts)→
More from Models
- ARC-AGI-3 leader changes: Yi-Chia Chen hits 59.17%, overtaking tufalabs — fchollet · 2026-10-09
- ARC-AGI-2 tops out at 88.06% as tufalabs claims ARC Prize 2026 high score — fchollet · 2026-10-09
- OpenAI's math gains likely from massive Lean-based RL environments; post-training is "rich man's inference" — yacineMTB · 2026-10-09
- Kalshi and Polymarket odds have Claude and Gemini neck-and-neck for best model by EOY on Argon 4 release — matt_slotnick · 2026-10-09
- Codex /fast mode mocked: 16-44 token/s vs Claude Opus 5.5's 80 — 'rename it /wait' — lxfater · 2026-10-09
- Power User Burns Five 20x Claude Plans a Week, Sparking Usage Debate — ChanceKelch · 2026-10-09