AI & Media
Text embeddings
Turn text into a vector for search or clustering. Priced per request.
Spot price
$0.0032per request
call-weighted across 23 of 27 priced endpoints
- Spread
- 10×
- cheapest to dearest
- Providers
- 18
- 19 hosts
- Endpoints
- 27
- 25 answered last probe
Where the offers sitLog scale. The shaded band is the middle half of the market, $0.0020 to $0.0050.
$0.0010$0.010
Every supplier below carries its index against that rate: 100 is spot, 40 is 60% below it, 310 is 3.1× it. Spot is weighted by how often each endpoint is called, so one wallet listing forty near-identical endpoints cannot move it on listings alone. The cheapest offer here is $0.0010 and the dearest $0.010.
Canonical request
| input | string | Text to embed, or a list of texts. |
| model? | string | Embedding model identifier. |
Canonical response
| embeddings | object[] | One vector per input. |
| embeddings[].vector | string[] | The vector components. |
| dimensions | number | Vector length. |
| model? | string | Model that produced them. |
2 models reachable over x402
Model families, not versions — publishers rarely pin one, and a gateway lets the caller choose at request time.| Model | Vendor | Sellers | Endpoints | Via gateway | Cheapest | Dearest |
|---|---|---|---|---|---|---|
| OpenAI embeddings | OpenAI | 5 | 8 | 4 | $0.0010 | $0.010 |
| Qwen | Alibaba | 1 | 1 | 0 | $0.0030 | $0.0030 |
Prices are per request, not per token. Of the chat-completion offers in this catalog, 66 state a per-call price and 3 mention per-token pricing at all — one seller’s call is ~128 tokens and the next leaves it unstated, so a $/1M-token comparison would be a number Roundhouse invented rather than one a publisher declared.
27 endpoints offering this capability, from 18 wallets
| Endpoint | Index | Models | Last called | |||||
|---|---|---|---|---|---|---|---|---|
x402 inference gateway | agent-production-1daa.up.railway.app/v1/embeddings | $0.0010 | 32 | 5 | 1 | GatewayOpenAI embeddings | Live | 46d ago |
 api.zuluworksai.com | api.zuluworksai.com/cf/embed | $0.0010 | 32 | 2 | 2 | — | Offline | 53d ago |
AiSpace Venice | x402.aispace.bot/api/v1/embeddings | $0.0010 | 32 | 10 | 3 | — | Live | 1d ago |
 x402engine.app | x402engine.app/api/embeddings | $0.0010 | 32 | 1 | 1 | OpenAI embeddings | Live | 6d ago |
 Agent402.tools | agent402.tools/v1/embeddings | $0.0020 | 63 | 58 | 2 | GatewayOpenAI embeddings | Live | today |
text-embeddings-vector-api | obol-x402.fly.dev/v1/ai/embed | $0.0020 | 63 | — | — | — | Offline | 36d ago |
 Quartermaster Embeddings | quartermaster.surewhynot.app/v1/embed | $0.0020 | 63 | 1 | 1 | — | Live | 12d ago |
Busker Embeddings | workbot1.oddsys.org/embed | $0.0020 | 63 | 1 | 1 | — | Live | 31d ago |
x402.forgemesh.io | x402.forgemesh.io/embeddings | $0.0020 | 63 | 1 | 1 | — | Live | 46d ago |
OpenAI-Compatible Embeddings | x402.shizu.me/v1/embeddings | $0.0020 | 63 | 1 | 1 | Gateway | Live | 27d ago |
 GEDX402 Knowledge | embed.gedx402.com/v1/embed/:model | $0.0030 | 95 | 8 | 1 | — | Live | 75d ago |
GEDX402 Knowledge | ged-x402-embed.jvalamis.workers.dev/v1/embed/:model | $0.0030 | 95 | 8 | 1 | Qwen | Live | 75d ago |
 402utils.com | 402utils.com/v1/embed | $0.0040 | 126 | 9 | 2 | — | Live | 44d ago |
 Text embeddings | agent402.tools/api/embed | $0.0050 | 158 | 6 | 2 | OpenAI embeddings | Live | today |
api.cortexcloud.org | api.cortexcloud.org/x402/v1/embeddings | $0.0050 | 158 | — | — | Gateway | Live | 28d ago |
 NetIntel | netintel.dev/api/v1/embeddings | $0.0050 | 158 | 1 | 1 | GatewayOpenAI embeddings | Live | 10d ago |
NetIntel | netintel.dev/v1/embeddings | $0.0050 | 158 | 1 | 1 | GatewayOpenAI embeddings | Live | 10d ago |
Embedding Compare API | orbisapi.com/proxy/embedding-compare-api-44b9f8 | $0.0050 | 158 | 1 | 1 | — | Live | 100d ago |
Embedding Normalizer API | orbisapi.com/proxy/embedding-normalizer-api-b765b8 | $0.0050 | 158 | 1 | 1 | — | Live | 100d ago |
Embedding Similarity API | orbisapi.com/proxy/embedding-similarity-api-985bc8 | $0.0050 | 158 | 2 | 1 | — | Live | 79d ago |
Text Embedding Search | orbisapi.com/proxy/text-embedding-search-578063 | $0.0050 | 158 | 1 | 1 | OpenAI embeddings | Live | 100d ago |
GEDX402 Knowledge | embed.gedx402.com/v1/embed | $0.0073 | 230 | 2 | 1 | — | Live | 75d ago |
GEDX402 Knowledge | ged-x402-embed.jvalamis.workers.dev/v1/embed | $0.0073 | 230 | — | — | — | Live | — |
Ollama API | ollama.bip-rep.com/v1/embeddings | $0.0080 | 252 | — | — | Gateway | Live | — |
Text embeddings (Large) | agent402.tools/api/embed-large | $0.010 | 315 | 6 | 2 | OpenAI embeddings | Live | today |
Reading this table
- Price is what the catalog carries for one unit, per request. It is a published figure, not a quote — read the live 402 before paying.
- Index is that price against the spot rate, so it moves only with price — there is no separate sort for it, because ordering by index and ordering by price are the same ordering.
- 30d calls and payers are the upstream catalog’s own counters for that endpoint, so they measure demand, not quality.
- Status is the last liveness probe. There is no latency or uptime column yet: a single probe outcome cannot support one, and inventing a percentage from it would be worse than leaving it out.
- Settled volume per endpoint is also absent, and cannot be derived — a settlement carries the payee wallet, not the endpoint that was called, so a wallet serving forty endpoints cannot be split between them. The provider wallet’s own record is one click away on each row.
- The canonical response above is the shape Roundhouse maps each provider onto. Field mappings per provider are not published yet; today the contract is the target, not a guarantee about any one endpoint’s payload.
Hand this page to an agent
Copy the prompt and paste it into Claude, an MCP client or your own agent — it will pick the best provider of this capability over the free read API. No key, no account.
GET www.roundhouse.studio/unified.json?capability=embeddings
Show the promptHide the prompt
Using Roundhouse, compare every x402 provider of Text embeddings and pick one for me. Weigh the price per unit against whether the endpoint answered when last probed, and say what the cheapest option would cost against the median. curl -s 'https://www.roundhouse.studio/unified.json?capability=embeddings' The canonical request and response in that document are the shape every provider is mapped onto. Before paying, check the provider's record and read the live 402: curl -s 'https://www.roundhouse.studio/api/v0/merchants/<the payTo wallet on the offer you choose>' If you do not have Roundhouse tools or skills installed, read https://www.roundhouse.studio/skill.md first — it is the whole procedure, and it needs no key.