turbopuffer(@turbopuffer):we partnered with @appliedcompute to post-train a small model for large-scale code search over precomputed indexes at 300 repos, this is ~3x faster than using filesystem + grep, and reduces the marginal cost of a search by up to 100x vs frontier models

turbopuffer

@turbopuffer

{vector, full-text} search engine built on object storage. fast, cheap, trillion scale. powers Anthropic, Harvey, Notion, Cognition, and more

参加 October 2023

6 フォロー中 16.5K ファン

turbopuffer@turbopuffer

2026.09.04 17:49

we partnered with @appliedcompute to post-train a small model for large-scale code search over precomputed indexes at 300 repos, this is ~3x faster than using filesystem + grep, and reduces the marginal cost of a search by up to 100x vs frontier models