tweetindex

DeepL @DeepLcom · 31 Aug 2026

There’s an inconvenient truth that nobody in AI translation likes to talk about. What we measure in quality test scores is becoming less important than what we don’t. Gaps in MQM-type quality benchmarks may be narrowing, but the differences in real-world performance are becoming https://t.co/BlZuMXFv4J
730Views
2Likes
1Reposts
0Replies
0Quotes
0Bookmarks

Is that a lot?

0.67×vs this author's median1 097 views is typical
0Percentile for this authorof 8 recent posts
0.58×vs 10K–100K median1 266 views is typical
3.66%Reachviews ÷ followers
0.41%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →