tweetindex

turbopuffer @turbopuffer · 13 Aug 2026

Coding agents issue dozens of context-fetch calls per task across multi-file, evolving repos. Most embedding models were never trained for that. 📢 Introducing voyage-code-4 — trained on retrieval patterns from real issue-fixing pull requests. 32K context. 📈voyage-code-4 https://t.co/Ceq17fvFxT
7 810Views
25Likes
4Reposts
2Replies
3Quotes
0Bookmarks

Is that a lot?

1.65×vs this author's median4 737 views is typical
60Percentile for this authorof 5 recent posts
2.70×vs 10K–100K median2 896 views is typical
50.10%Reachviews ÷ followers
0.43%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →