Sakana AI's Namazu translation model claims to beat Google Translate and DeepL
SakanaAILabs · x · 2026-08-22
Sakana AI upgraded its translation service Sakana Translate to the new-generation Sakana Namazu model, maintaining bidirectional Japanese-English-Chinese translation while improving output naturalness.
On its own 160-task Japanese-English benchmark, Sakana claims its output was judged better than all comparison targets — including Google Translate and DeepL — in over 50% of cases. Note this is a vendor-run evaluation, not an independent benchmark.
Related event: Sakana AI Upgrades Sakana Translate with New Namazu Model(5 posts)→
More from Models
- Grok Bot Ran a 12-Hour Data Analysis Nonstop, Handling API Rate Limits Itself — Daniel_Farinax · 2026-10-16
- One year of Recursive Language Models: GPT-5-mini beats GPT-5 by 110% on 132k-token tasks, cheaper too — a1zhang · 2026-10-16
- Users Report Claude Models Suddenly More Compliant After Forced Compaction — RealSharpNinja · 2026-10-16
- Prediction: Opus 5.5-level models on laptops by 2027, backed by Qwen 27B data — cephaloform · 2026-10-16
- Users Say Krea 2 Output Degraded Sharply Since Krea 3 Beta Launched — szywis · 2026-10-16
- LLM-as-a-Verifier lifts Nemotron 3.5 Super by 9-10 points on agentic benchmarks — simonguozirui · 2026-10-16