Comparison · 8 min · 4 citations
Weaviate vs Qdrant Pricing in 2026
Weaviate vs Qdrant pricing in 2026: per-dimension serverless billing against fixed-cluster cost, with current rates, free tiers and which wins small.
Both vendors now run free-forever tiers, and they cap different things. Weaviate's Free Forever cluster holds 100,000 objects in one collection with 1 GB memory and 10 GB disk[2]; Qdrant's free cluster has 0.5 vCPU, 1 GB RAM and 4 GB disk with no object cap[3]. A small RAG app under 100,000 vectors can run free on either; past that, Weaviate's paid entry is the Flex tier at $45 per month[2].
The paid tiers bill on different axes. Weaviate Cloud charges per million vector dimensions per month, from $0.00465 on Flex and $0.003875 on Premium[1], so cost scales with vector count times embedding dimension. Qdrant charges for the compute and memory a cluster reserves, with exact per-GB-hour rates not publicly documented[3]. At 200,000 1,536-dimension vectors, Weaviate's dimension fee is only about $1.43 per month, so the $45 Flex minimum is the number that actually applies.
Verified as of 2026-10-18 against the official vendor pricing and documentation pages cited below.
The choice between the two turns on billing axis once you outgrow the free tiers: Weaviate charges per million stored vector dimensions with a monthly minimum, while Qdrant charges for reserved cluster compute. A 1,536-dimension model costs twice as much per vector on Weaviate as a 768-dimension model, because the bill scales with dimensions. This article runs the Weaviate per-dimension math on a concrete scenario and lays out the verified rates against Qdrant.
1. Per-dimension vs per-cluster: two billing axes
Weaviate Cloud's serverless model bills on stored vector dimensions: from $0.00465 per million dimensions per month on Flex, $0.003875 on Premium and $0.002718 on Enterprise, with storage from $0.12 per GiB on Flex[1]. Weaviate notes that dimension pricing varies by index type, compression method and region[1]. Total dimensions equal vector count times embedding dimension, so the same vector count costs twice as much at 1,536 dimensions as at 768. Flex carries a 99.5% SLA and a $45 per month minimum; Premium starts at $400 per month with a 99.9% SLA, and Enterprise reaches 99.95%[1][2].
Qdrant Cloud bills on reserved cluster resources (vCPU, RAM, disk), hourly, with a 99.5% SLA on its usage-based Standard tier and 99.9% on Premium, which carries a minimum spend[3]. Dimension count affects only how much fits in a given cluster's memory, not a per-dimension line item. The exact paid per-GB-hour rate is not published on the official pricing page as of October 2026; Qdrant directs users to its calculator. The two models reward opposite workloads: Weaviate is predictable when dimensions are stable, Qdrant is efficient when a cluster runs busy.
2. Weaviate per-dimension math on the scenario
Scenario: 200,000 vectors at 1,536 dimensions. Total stored dimensions are 200,000 times 1,536, which is 307.2 million dimensions. The formula for the Weaviate dimension fee is total-million-dimensions times the per-million rate:
Flex dimension fee = 307.2 million dims × $0.00465 / million = $1.43 / month
Premium dimension fee = 307.2 million dims × $0.003875 / million = $1.19 / month
The dimension fee at this scale is tiny. Adding about 1.3 GiB of storage at $0.12 per GiB still leaves the metered total under $2, so what actually applies is the Flex tier's $45 per month minimum[1][2]. The dimension model only becomes the dominant cost at large vector counts: a deployment of 5 million 1,536-dimension vectors with a replication factor of 2 reaches roughly $71 per month in dimension fees alone (7,680 million dimensions times $0.00465 times 2 replicas), still under Premium's $400 minimum.
For storage-footprint reference at the same scenario, the Embeddings DB Cost engine reports the raw stored bytes (this engine prices Pinecone, pgvector, LanceDB, and Turbopuffer at current rates with plan minimums, not Weaviate or Qdrant, so the footprint line is the part that carries over to this pair):
Show the recompute-verified inputs and outputs
| vector_count | 200000 |
|---|---|
| dim | 1536 |
| queries_per_day | 2400 |
| ingest_per_day | 1500 |
| vendors › row 1 › vendor | Pinecone |
|---|---|
| vendors › row 1 › monthly cost | 50 |
| vendors › row 1 › notes | Pinecone Standard 2026-10: $16/M read units, $4/M write units, $0.33/GB-mo, $50/mo plan minimum; a query here uses about 1.43 read units (1 per GB of namespace, minimum 0.25). Small workloads can fit the free Starter plan (up to 2 GB, 1M read and 2M write units a month) or Builder ($20/mo flat, up to 10 GB, 2M read and 5M write units). |
| vendors › row 2 › vendor | Postgres+pgvector |
| vendors › row 2 › monthly cost | 125 |
| vendors › row 2 › notes | Supabase Pro 2026-10: $25/mo with $10 compute credit, Large compute ($110/mo, 8 GB RAM), disk above 8 GB at $0.125/GB-mo for about 2.86 GB (table plus HNSW index). Sized from Supabase's pgvector HNSW benchmarks (1536-d: Micro 15k vectors, Small 50k, Medium 100k, Large 224k, XL 500k, 2XL 1M), scaled by floats per vector and by RAM above 2XL. |
| vendors › row 3 › vendor | LanceDB |
| vendors › row 3 › monthly cost | 0.25 |
| vendors › row 3 › notes | LanceDB on Cloudflare R2 2026-10: $0.015/GB-mo storage, $4.50/M writes (Class A), $0.36/M reads (Class B), one request per upsert and per query. Self-hosted compute not included. |
| vendors › row 4 › vendor | Turbopuffer |
| vendors › row 4 › monthly cost | 16 |
| vendors › row 4 › notes | Turbopuffer Launch 2026-10: $16/mo minimum usage; $0.33/GB-mo logical storage, $2/GB written, $0.001/TB of data queried (namespace size, 1.28 GB minimum, discounted above 32 GB), $0.05/GB returned. |
| cheapest vendor | LanceDB |
| cheapest monthly cost | 0.25 |
| cheapest all in vendor | Turbopuffer |
| cheapest all in monthly cost | 16 |
| storage gb | 1.43 |
Computed live at build time.
The engine reports a storage footprint near 1.43 GB at this scale. That fits Weaviate's 10 GB free disk, but 200,000 objects exceed its 100,000-object free cap. It also fits Qdrant's 4 GB free disk, though 200,000 raw 1,536-dimension float vectors need about 1.2 GB, more than the free cluster's 1 GB of RAM, so you would keep vectors on disk or quantize them to stay free.
3. Free tiers compared
Weaviate's Free Forever plan is a managed cluster with 100,000 objects, 1 GB memory and 10 GB disk, one collection with up to 3 tenants, embeddings at 2,000 requests a day and no credit card required[2]. Qdrant offers a free-forever single-node cluster with 0.5 vCPU, 1 GB RAM and 4 GB disk[3].
Neither expires, so the distinction is what each caps. Weaviate limits the object count and the number of collections; Qdrant limits the hardware. A side project with one collection under 100,000 vectors runs free on either. A larger single index, or an app that needs several collections, fits Qdrant's free tier if its vectors fit the resources, and needs Weaviate's $45 Flex tier.
4. The verified comparison table
| Dimension | Weaviate Cloud | Qdrant Cloud |
|---|---|---|
| Billing axis | Per million stored dimensions[1] | Per cluster compute/memory/disk[3] |
| Entry tier price | $45 / month (Flex)[2] | None (pay for cluster) |
| Dimension rate (Flex) | From $0.00465 / M dims/mo[1] | Bundled into cluster |
| Premium tier | From $400 / month, 99.9%[2] | Minimum spend, 99.9%[3] |
| Entry paid SLA | 99.5% (Flex)[1] | 99.5% (Standard)[3] |
| Free tier | Free Forever: 100,000 objects, 1 GB memory, 10 GB disk[2] | Free-forever cluster: 0.5 vCPU, 1 GB RAM, 4 GB disk[3] |
| Exact paid per-unit rate | Published[1] | Not publicly documented (calculator)[3] |
5. Which to pick
- Pick Qdrant if you want a free cluster without an object cap, predictable fixed-cost compute, or the option to self-host the identical engine later. Below the free-tier resource ceiling the cost is zero.
- Pick Weaviate if you want per-dimension billing that scales smoothly with no cluster to size, published SLA tiers, and a hybrid-search feature set. It is free up to 100,000 objects in one collection; beyond that, budget the $45 Flex minimum.
- Watch the dimension count. Weaviate's per-dimension model penalizes high-dimension embeddings. If you use 1,536-dimension or larger vectors at volume, model the dimension fee carefully before committing.
For the full landscape, the cheapest vector database ranking prices eight vendors at one scenario, and the Pinecone vs Qdrant article covers the per-usage alternative to both.
Frequently asked questions
Is Weaviate or Qdrant cheaper in 2026?
Both now have free-forever tiers. Weaviate's holds 100,000 objects in one collection (1 GB memory, 10 GB disk); Qdrant's is a 0.5 vCPU, 1 GB RAM, 4 GB disk cluster with no object cap. Above those, Weaviate's Flex tier starts at $45 per month and bills from $0.00465 per million vector dimensions, so at 200,000 1,536-dimension vectors the $45 minimum is what you pay, while Qdrant bills for the compute and memory a cluster reserves.
How does Weaviate Cloud serverless pricing work?
Weaviate Cloud bills per million vector dimensions stored per month, from $0.00465 per million on Flex (99.5% SLA, $45 per month minimum), $0.003875 on Premium (99.9%, from $400 per month) and $0.002718 on Enterprise (99.95%), plus storage from $0.12 per GiB on Flex. Dimension rates vary by index type, compression and region. Total dimensions equal vector count times the embedding dimension.
Does Weaviate have a free tier?
Yes. Weaviate's Free Forever plan is a managed cluster with 100,000 objects, 1 GB memory and 10 GB disk in one collection (up to 3 tenants), with no credit card required. Qdrant's free-forever cluster has 0.5 vCPU, 1 GB RAM and 4 GB disk. Weaviate caps the object count; Qdrant caps the resources (checked 2026-10-18).
References
Sources
Primary sources only. No vendor-marketing blogs or aggregated secondary claims.
- 1 Weaviate — Pricing dimensions (vector dimensions from $0.00465 per million on Flex, $0.003875 Premium, $0.002718 Enterprise; storage, backups, SLAs) — accessed 2026-10-18
- 2 Weaviate — Plans (Free Forever: 100,000 objects, 1 GB memory, 10 GB disk; Flex from $45 a month; Premium from $400 a month) — accessed 2026-10-18
- 3 Qdrant — Pricing (free-forever 0.5 vCPU / 1 GB RAM / 4 GB disk; usage-based Standard tier at 99.5%; Premium with minimum spend at 99.9%) — accessed 2026-10-18
- 4 AI Biz Hub — Embeddings DB Cost methodology — accessed 2026-10-18
The 9 vendor figures on this page, each with its source and the date it was checked, are also in the vendor price reference, where every figure has its own link.
Tools referenced in this article
Related articles