At a glance — Oct 2026
Updated 2026-10-27 · forward this page: /jtbds/prototype-rag#briefScores as of 2026-09-14 (43 days old; Depot marks scores stale after 30 days). The pick is still Depot's current editorial pick; it has not been re-reviewed since. Prices below carry their own dates.
- Depot Pick
- Chroma · 7.88/10 for this job (curated estimate, high confidence)
dx first; reliability discounted.
- Runner-up
- Supabase (pgvector) · 8.64/10
Scores higher than the pick on: Capability fit, Latency / reliability, Developer & agent DX, Privacy / data retention, Maturity / track record.
- When not to use the pick
- Production traffic — graduate immediately.
- Lock-in / migration
- The pick has lower stored lock-in risk than runner-up Supabase (pgvector) (2.5 vs 3). Pick exit signals: open source (provider tag `oss`). Self-host fallback listed for this job: Chroma.
- Sourced price (no comparable unit)
Pick: no sourced price in this job's unit (no_sourced_price).
No cheapest SKU (no_comparable_unit_for_job). No price was invented.
- Free option
- Chroma (oss) — source, as of 2026-10-25
- Failure modes to watch
- Context overflow: If the input exceeds the window, say so and retrieve a narrower slice. Do not answer from a truncated head as if it were the whole source.
- Eval contamination: Keep gold labels out of the agent prompt. Store the harness version beside the score. Do not compare a number across a benchmark revision without saying the revision changed.
- Lost in the middle: Ask the agent to name the passage it used. Spot-check questions whose only support sits in the middle of the context, not only at the edges.
- Policy conflict: If cited sources contradict each other, return the conflict and the URLs. Do not pick a winner silently.
- Silent truncation: Tools must return whether the payload was truncated and how long the original was. Fail the task if the answer depends on the missing span.
- Stale retrieval: Store the source URL and the time it was fetched. If a newer canonical page exists, or two stored sources disagree, do not answer from the older one.
Cite: Agent Provider Depot, Depot Pick for prototype-rag: chroma, scores as of 2026-09-14. Agents: GET /api/decide?jtbd=prototype-rag
Free Prototype RAG in a day options · for rag agents
Chroma
dx first; reliability discounted.
Pricing (directional): As of 2026-09-14: Open-source embeddable vector DB + hosted Chroma Cloud; free/local for prototypes; Cloud usage-based for managed. Excellent DX for demos; graduate before high QPS/filtered production.
Lock-in risk: higher = more lock-in (worse). Overall already inverts this dimension.
Prototype on Chroma; graduate to Qdrant/pgvector/Pinecone before your corpus and QPS grow up.
Quote for Prototype RAG in a day
Depot Pick stays Chroma. Prices below are sourced SKU rows only (as of 2026-10-27). Input and output are not blended. GET /api/quote?jtbd=prototype-rag
Nothing was compared. no comparable sourced unit for this job; chat input $/1M and other units were not borrowed pickIsPriceWinner is null.
Sourced price unavailable for the Depot Pick. No row in the pricing index.
7 of 7 candidates have no sourced price row. 7 are excluded from the Matching price comparison (none). no comparable sourced unit for this job; chat input $/1M and other units were not borrowed
Runners-up
Supabase (pgvector)
Depot default for most agent apps under ~25–50M vectors. Don't add Pinecone until Postgres actually hurts.
Pinecone
Still the zero-ops vector default. If you already run Postgres, try pgvector before paying Pinecone rents.
Upstash Vector
Prototype/edge RAG peer when you already on Upstash. Prefer Qdrant/Pinecone for serious filtered production…
When NOT to use the top pick
Production traffic — graduate immediately.
Full ranking (score table)
Ranked with this JTBD's weights. Scores are curated estimates unless marked otherwise. Agents: prefer /api/decide then this table for tie-breaks.
| # | Provider | Overall | Fit | $/perf | Latency | DX | As of |
|---|---|---|---|---|---|---|---|
| 1 | Supabase (pgvector) | 8.6 | 8.5 | 9.0 | 8.6 | 9.4 | 2026-09-14 |
| 2 | Upstash Vector | 8.2 | 7.8 | 9.0 | 8.4 | 9.0 | 2026-09-14 |
| 3 | DuckDB (vector/extensions) | 8.0 | 6.0 | 9.5 | 7.5 | 8.5 | 2026-09-14 |
| 4 | Pinecone | 8.0 | 9.0 | 6.2 | 9.0 | 8.8 | 2026-09-14 |
| 5 | ChromaDepot Pick | 7.9 | 7.2 | 9.0 | 7.0 | 9.2 | 2026-09-14 |
| 6 | LanceDB | 7.8 | 7.8 | 8.8 | 7.2 | 8.2 | 2026-09-14 |
| 7 | Mem0 | 7.6 | 7.8 | 8.0 | 7.4 | 8.8 | 2026-09-14 |
Score methodology for this job
dx first; reliability discounted.
capabilityFit: 25%pricePerformance: 18%latencyReliability: 15%developerDx: 15%privacyPosture: 10%lockInRisk: 7%maturity: 10%