Turbopuffer on X / Twitter

Official accounts (1)

turbopuffer @turbopuffer · Company Primary

Active Accounts: 1 tracked
Posts YTD: 80 ↑ vs 0 last year
Avg Posts/Month: 6.7 past 12 months
Views: 2,062,304 80/80 reported
Likes: 12,256 80/80 reported
Reposts: 1,066 80/80 reported
Replies: 468 80/80 reported
Quotes: 320 80/80 reported
Bookmarks: 5,237 80/80 reported

Monthly Activity

Account: Date range:

Engagement values are the latest totals observed for each post, grouped by publication month. Provider coverage can vary by metric; unavailable values are shown as --.

Post Details

Search:

Post	Account	Account Type	Post Type	Published	Views	Likes	Reposts	Replies	Quotes	Bookmarks
RT @meganreyno: Vibes were immaculate at the https://t.co/u3TjotHZgu systems meet-up x @turbopuffer tonight ✨ Packed …	@turbopuffer	Company	Repost	2026-10-29	88	0	3	0	0	0
RT @softwaredoug: Tomorrow I'm continuing my discussion of agent memory with Turbopuffer. Join me! Indexing your corpus…	@turbopuffer	Company	Repost	2026-10-28	79	0	1	0	0	0
we’re serving up scoops of puffmallow in SF with @JudgmentLabs to celebrate their GA come grab a cup or cone at 1 Mont…	@turbopuffer	Company	Original	2026-10-28	2,346	34	3	4	1	2
we've been using Claude Mythos for our internal defense systems we could ask Mythos to examine every log source, but t…	@turbopuffer	Company	Quote	2026-10-28	23,554	375	12	9	1	313
RT @colossusmag: The people of Silicon Valley are unwell. If you work at a frontier lab or decacorn, you’ve got it bad…	@turbopuffer	Company	Repost	2026-10-28	85	0	46	0	0	0
turbopuffer v3 will be the first search engine to adopt fully vectorized query execution, a design more often used by O…	@turbopuffer	Company	Original	2026-10-26	26,650	306	13	7	6	140
RT @bentlegen: in hindsight we should have switched to @turbopuffer much earlier https://t.co/6B9FHZ2shL	@turbopuffer	Company	Repost	2026-10-23	91	0	5	0	0	0
the official tpuf Rust SDK is now in beta 🦀 typed, async access to tpuf from your Rust applications https://t.co/XWg5…	@turbopuffer	Company	Original	2026-10-23	4,776	74	2	2	2	16
making search fast means touching fewer bytes and doing less work per byte tpuf v3 compresses columns with composable …	@turbopuffer	Company	Original	2026-10-22	3,401	106	3	0	1	31
RT @harvey: Review Tables now load up to 5.8x faster in Harvey. Our engineering team built server-side pagination to s…	@turbopuffer	Company	Repost	2026-10-22	81	0	2	0	0	0
Nemotron-3.5-Embed is now available in turbopuffer native embedding on our enterprise PDF search benchmark, Nemotron-3…	@turbopuffer	Company	Quote	2026-10-21	3,306	77	1	5	2	25
native reranking is generally available retrieve first-stage candidates and reorder them with a reranker model in a si…	@turbopuffer	Company	Original	2026-10-20	4,938	72	3	4	2	19
rescoring fewer full-precision vectors cut disk read I/O by up to 80% on one prod cluster lower disk I/O produces high…	@turbopuffer	Company	Quote	2026-10-19	11,889	154	6	2	4	30
RT @KaiLentit: Hyperoptimized Engineer. https://t.co/rel0iQu534	@turbopuffer	Company	Repost	2026-10-16	95	0	374	0	0	0
a great way to make search fast is to skip work our basic unfiltered order by on tpuf v3 did a full scan. we cut laten…	@turbopuffer	Company	Original	2026-10-09	4,341	79	2	4	2	21
filters on tpuf v3 are now faster than prod (v2) instead of filtering row-at-a-time, v3 will push compute down to stor…	@turbopuffer	Company	Original	2026-10-08	5,247	55	3	0	1	9
binary quantization lets tpuf store & score more vectors in RAM. to keep recall high, we rescore full-precision can…	@turbopuffer	Company	Original	2026-10-08	9,269	100	5	8	6	25
september changelog https://t.co/GcMovQKbAN https://t.co/2Ou5xAwpwr	@turbopuffer	Company	Original	2026-10-07	1,567	22	1	1	0	1
tpuf v3's new storage layout will enable batched execution on all query types — including filters, order by, and aggreg…	@turbopuffer	Company	Original	2026-10-06	2,745	54	2	1	1	10
we implemented batched reads in the v3 full-text engine, leading to a ~9x performance improvement. postings are stored …	@turbopuffer	Company	Original	2026-10-02	42,131	168	6	1	2	70
since v1, tpuf has keyed every index on the vector address we've stretched this architecture to its limit. to make sea…	@turbopuffer	Company	Original	2026-10-01	4,577	87	4	19	1	23
we maintain several internal benchmarks to guide our customers toward better search relevance @perplexity's new model…	@turbopuffer	Company	Quote	2026-09-30	32,682	152	17	7	1	59
we're changing tpuf's storage architecture so we can keep pushing the search frontier the new engine (v3) now passes C…	@turbopuffer	Company	Original	2026-09-30	83,681	339	18	22	11	132
moving embedding into turbopuffer lets us parallelize the embedding call with other work in the query plan whole-syste…	@turbopuffer	Company	Original	2026-09-29	20,531	105	0	0	6	39
native reranking is now in private beta rerank query results using state-of-the-art reranker models, including Jev, wi…	@turbopuffer	Company	Quote	2026-09-24	45,579	369	15	17	3	295
RT @wilsonsaccount: don’t talk to me until i’ve squished my squishypuffs https://t.co/X2HQ5QkeNG	@turbopuffer	Company	Repost	2026-09-23	83	0	1	0	0	0
ask Kepler for a public company's revenue growth, and it traces the number to an exact line item in a verified SEC fili…	@turbopuffer	Company	Original	2026-09-23	134,406	57	7	4	6	32
new: bytes data type before, you stored binary data in tpuf as a string and paid ~33% extra for encoding. now, you pay…	@turbopuffer	Company	Original	2026-09-18	5,060	80	4	1	0	19
august changelog https://t.co/GcMovQKbAN https://t.co/rFDhj48FZ2	@turbopuffer	Company	Original	2026-09-15	2,961	25	0	0	1	5
tpuf native embeddings are now generally available embed documents and queries without another third-party API call	@turbopuffer	Company	Quote	2026-09-14	23,577	196	7	5	4	114
RT @DevinLewtan: we also couldn’t help but make a small batch of limited edition cowpuffs! now live on https://t.co/8C…	@turbopuffer	Company	Repost	2026-09-11	91	0	5	0	0	0
new: explore namespaces in the tpuf dashboard → browse all documents in your namespaces → check metadata and indexing …	@turbopuffer	Company	Original	2026-09-11	32,457	153	5	9	3	91
your product is only as smart as the data it can reach. make your product smart. unleash your cow. https://t.co/Uck61X…	@turbopuffer	Company	Original	2026-09-10	328,815	1,502	77	81	97	659
in the last 30 days, the p100 full-text query on tpuf evaluated 1,823 clauses agents write (much) longer queries than …	@turbopuffer	Company	Original	2026-09-09	5,647	49	0	5	2	11
we partnered with @appliedcompute to post-train a small model for large-scale code search over precomputed indexes at …	@turbopuffer	Company	Original	2026-09-04	41,134	436	39	10	8	289
Entire searches 450M+ code blocks, prompts, and agent session transcripts on turbopuffer → namespace per repository → …	@turbopuffer	Company	Quote	2026-09-03	18,895	191	16	1	1	99
fal just made it possible to generate video faster than you can watch it turbopuffer makes it possible to search every…	@turbopuffer	Company	Original	2026-09-01	16,669	132	17	7	3	75
new: multi-attribute ordering order query results by up to 8 attributes docs: https://t.co/GNAb40ikIB https://t.co/Fj…	@turbopuffer	Company	Original	2026-08-27	3,938	57	2	1	0	15
turbopuffer's first public Azure region, azure-eastus2, is now in private preview contact us for access https://t.co/k…	@turbopuffer	Company	Original	2026-08-20	5,208	88	1	5	1	12
voyage-code-4 is now available on tpuf native embeddings contact us to join the beta full model list here: https://t.…	@turbopuffer	Company	Quote	2026-08-19	5,250	22	0	4	0	12
when you open Linear, the client syncs changes since your last session Linear moved the delta sync read path from Post…	@turbopuffer	Company	Quote	2026-08-19	23,895	292	5	2	1	109
we run 100+ tpuf clusters, including many that live inside customers' clouds (BYOC). how do you operate a cluster you c…	@turbopuffer	Company	Original	2026-08-14	123,195	484	40	15	23	462
july changelog https://t.co/dYfZZ2mzg9 https://t.co/idQn3fyhSo	@turbopuffer	Company	Original	2026-08-11	3,317	33	2	0	0	7
sharding is now in beta index the world's largest corpora (up to 256 TB) in a single namespace docs: https://t.co/UlV…	@turbopuffer	Company	Quote	2026-08-05	35,472	322	18	3	1	227
Mem0 migrated 400M+ agent memories from pgvector to turbopuffer, solving a semi-selective filtering problem that spiked…	@turbopuffer	Company	Original	2026-07-30	6,570	130	10	2	2	44
tpuf now supports late interaction [beta] use models like ColBERT to represent text as a set of vectors (1 per token) …	@turbopuffer	Company	Original	2026-07-29	44,520	170	15	7	10	66
tpuf BM25 index tracks matching doc IDs per term: [1, 8, 32, ... 1048576] instead of IDs (big), we store the deltas (…	@turbopuffer	Company	Quote	2026-07-24	14,335	108	18	4	1	62
autoscaling is deceptively hard our indexer fleet scales nodes on job queue time. if a queue suddenly went quiet, we'd…	@turbopuffer	Company	Original	2026-07-22	27,121	170	8	7	7	53
new: computed attributes compute additional values on the results of the rank_by and pass them as extra signals to a r…	@turbopuffer	Company	Original	2026-07-20	6,814	69	1	0	2	29
new: highlighting extract the text fragments most relevant to a query → highlight matches in your search results UI →…	@turbopuffer	Company	Original	2026-07-17	10,183	194	8	3	1	91
Nemotron 3 Embed is now available on tpuf native embeddings via our partner @baseten contact us for beta access, full…	@turbopuffer	Company	Quote	2026-07-16	7,488	76	3	1	0	23
june changelog https://t.co/dYfZZ2mzg9 https://t.co/Pn17uEi7KR	@turbopuffer	Company	Original	2026-07-15	4,118	53	0	1	0	12
now in beta: native embeddings in tpuf embedding is the most painful part of puffing. we want to make it easy you can…	@turbopuffer	Company	Original	2026-07-14	48,575	238	12	12	8	79
TP SUMMER 26: the beach puff 🌞 now live on https://t.co/SsVQ3P6HeK https://t.co/iH0hAf1jdQ	@turbopuffer	Company	Original	2026-07-07	17,459	54	2	3	2	8
Legora searches 2B+ legal documents on turbopuffer → strict per-matter data isolation with CMEK → 10x lower tail laten…	@turbopuffer	Company	Original	2026-06-30	12,460	101	9	3	2	62
new: i8 vectors f32: 4 bytes/dim i8: 1 byte/dim 4x fewer bytes → 75% lower storage and query costs + faster queries …	@turbopuffer	Company	Original	2026-06-18	31,735	162	11	2	2	54
we cut our base price from $64 → $16/mo start puffing for 4x less https://t.co/1R8OrNfiKF	@turbopuffer	Company	Original	2026-06-18	42,799	465	16	24	8	100
we open sourced alyze, the Rust crate behind tpuf's default full-text search tokenizer (word_v4) our first tokenizers …	@turbopuffer	Company	Original	2026-06-17	14,335	230	15	3	3	121
Atlassian's cross-product AI platform, Rovo, searches 5B+ documents on turbopuffer BYOC → 19% increase in search quali…	@turbopuffer	Company	Original	2026-06-11	5,839	88	4	4	2	18
a year ago, ~98% of tpuf queries were vector ANN last 30d: 64% vector ANN 19% full-text BM25 13% filter-only 3% aggre…	@turbopuffer	Company	Original	2026-06-08	22,929	150	4	6	2	34
new: rerank_by before, you'd implement rank fusion client-side. now, a little QoL upgrade, especially nice for large r…	@turbopuffer	Company	Original	2026-06-05	5,528	81	3	6	0	25
may changelog https://t.co/dYfZZ2mzg9 https://t.co/ub7R8UhRwD	@turbopuffer	Company	Original	2026-06-03	2,331	27	0	1	0	1
new: branching create an instant, copy-on-write clone of a tpuf namespace → constant-time (440ms p50, ~1s p99) → full…	@turbopuffer	Company	Original	2026-06-02	30,065	140	8	8	3	38
tpuf quantizes vectors to improve perf (RaBitQ) the algo randomly rotates vectors, and we were using matmul at O(d²) s…	@turbopuffer	Company	Original	2026-06-01	11,500	149	10	6	2	46
new in turbopufer: the Fuzzy filter typo-tolerant substring matching with a configurable edit distance, so you can puf…	@turbopuffer	Company	Original	2026-05-22	46,973	131	5	8	4	50
puff	@turbopuffer	Company	Quote	2026-05-21	19,743	169	1	13	1	10
SID-1 is an agentic search model by @SID_AI → 1.9x recall over RAG + rerank → 24x faster, 99% cheaper than GPT-5.1 tr…	@turbopuffer	Company	Original	2026-05-20	114,454	218	26	7	8	205
filtered counts on tpuf just got much faster our ANN index may replicate docs across clusters for better recall, so we…	@turbopuffer	Company	Original	2026-05-13	7,802	64	3	1	1	19
we're bringing more of the database to the tpuf dashboard step one: namespace metadata more soon https://t.co/2BuId6h…	@turbopuffer	Company	Original	2026-05-12	5,503	50	0	1	0	14
new: sparse vectors a first-class retrieval primitive that composes with BM25 + attribute ranking in the same query pl…	@turbopuffer	Company	Original	2026-05-08	15,529	129	9	5	3	67
new: namespace pinning pin namespaces to reserved compute for faster, cheaper, more predictable p99 on high QPS worklo…	@turbopuffer	Company	Original	2026-05-07	18,224	71	6	5	3	23
april changelog https://t.co/dYfZZ2mzg9 https://t.co/vVCszLUG8K	@turbopuffer	Company	Original	2026-05-06	4,395	31	0	0	1	6
this spring, we've been having a lot of fun making some puffy inspired pieces for our customers who've been puffin' alo…	@turbopuffer	Company	Original	2026-05-06	10,709	52	3	5	3	11
new tpuf regions, so you can puff a little closer to home 🇧🇷 aws sa-east-1 (São Paulo) 🇺🇸 gcp us-east1 (South Carolina…	@turbopuffer	Company	Original	2026-05-04	5,253	54	1	4	1	6
⛷️ turbopuffer summit #004 🏂 our remote team spent a whole week together in banff national park. we brought a camcorde…	@turbopuffer	Company	Original	2026-04-30	45,866	234	6	10	7	43
✨ docs search for search docs ✨ now live on https://t.co/CKbgkiA39v https://t.co/Dt0r1tVPs2	@turbopuffer	Company	Original	2026-04-28	29,571	77	3	4	1	30
BM25 efficiently scores text, but relevance often depends on more than text (recency, popularity, PageRank) we score n…	@turbopuffer	Company	Original	2026-04-27	21,334	124	12	0	3	95
stemming is what makes a text search for "run" match documents containing "running" or "runs" we just shipped a small …	@turbopuffer	Company	Original	2026-04-21	4,583	51	0	2	1	9
march changelog https://t.co/dYfZZ2mzg9 https://t.co/T5wjr8Iuyi	@turbopuffer	Company	Original	2026-04-15	3,282	24	0	0	0	8
we couldn’t afford the sphere so we got the next best thing costs less, hits harder https://t.co/EgTpcfgQEv	@turbopuffer	Company	Original	2026-04-14	198,580	1,075	41	34	22	177

Plushcap, by Matt Makai. 2021-2026.