Skip to content
Get an API key

Live · serverless · 2 regions

Qwen3 Embedding 8B

by Alibaba

An 8B embedding model that ranks near the top of multilingual retrieval benchmarks and supports instructions in the query.

  • embeddings
  • BF16
Serverless priceUSD
Input
$0.02per 1M tokens
Output
—returns vectors, not tokens

Billed per token, no minimums.

Reference

Specifications

Model ID
qwen3-embedding-8b
Endpoint
POST /v1/embeddings
Publisher
Alibaba
Tasks
embeddings
Served context
32K tokens
Quantization
BF16
p50 response
41 mslast 24 hours
Regions
us-east · eu-west
Dedicated
On request

Last 24 hours

Latency

Median of each 15-minute step on the shared pool, measured at the API edge, for the whole response: there is no token stream to time.

Response time · ms

41 msmedian

p95 52 ms

Show the data
Median of each 15-minute step over the last 24 hours, newest first
StepResponse time
15m ago43 ms
30m ago44 ms
45m ago55 ms
1h ago42 ms
1h 15m ago41 ms
1h 30m ago44 ms
1h 45m ago44 ms
2h ago42 ms
2h 15m ago44 ms
2h 30m ago42 ms
2h 45m ago44 ms
3h ago41 ms
3h 15m ago43 ms
3h 30m ago44 ms
3h 45m ago43 ms
4h ago42 ms
4h 15m ago41 ms
4h 30m ago43 ms
4h 45m ago61 ms
5h ago41 ms
5h 15m ago42 ms
5h 30m ago42 ms
5h 45m ago40 ms
6h ago41 ms
6h 15m ago42 ms
6h 30m ago41 ms
6h 45m ago43 ms
7h ago42 ms
7h 15m ago44 ms
7h 30m ago41 ms
7h 45m ago41 ms
8h ago41 ms
8h 15m ago41 ms
8h 30m ago43 ms
8h 45m ago39 ms
9h ago40 ms
9h 15m ago41 ms
9h 30m ago41 ms
9h 45m ago41 ms
10h ago41 ms
10h 15m ago39 ms
10h 30m ago40 ms
10h 45m ago40 ms
11h ago39 ms
11h 15m ago41 ms
11h 30m ago39 ms
11h 45m ago40 ms
12h ago41 ms
12h 15m ago41 ms
12h 30m ago53 ms
12h 45m ago39 ms
13h ago39 ms
13h 15m ago39 ms
13h 30m ago39 ms
13h 45m ago39 ms
14h ago38 ms
14h 15m ago39 ms
14h 30m ago40 ms
14h 45m ago40 ms
15h ago40 ms
15h 15m ago41 ms
15h 30m ago41 ms
15h 45m ago54 ms
16h ago41 ms
16h 15m ago41 ms
16h 30m ago41 ms
16h 45m ago40 ms
17h ago52 ms
17h 15m ago41 ms
17h 30m ago41 ms
17h 45m ago39 ms
18h ago39 ms
18h 15m ago41 ms
18h 30m ago41 ms
18h 45m ago39 ms
19h ago39 ms
19h 15m ago41 ms
19h 30m ago40 ms
19h 45m ago42 ms
20h ago40 ms
20h 15m ago41 ms
20h 30m ago42 ms
20h 45m ago40 ms
21h ago42 ms
21h 15m ago40 ms
21h 30m ago41 ms
21h 45m ago41 ms
22h ago42 ms
22h 15m ago40 ms
22h 30m ago40 ms
22h 45m ago40 ms
23h ago43 ms
23h 15m ago41 ms
23h 30m ago44 ms
23h 45m ago42 ms
24h ago42 ms

API

Make a request

The same request shape as every other model on /v1/embeddings. Switching between them is a change to one string.

curl https://api.weir.example/v1/embeddings \
  -H "Authorization: Bearer $WEIR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-embedding-8b",
    "input": "Where is my order?"
  }'
POST /v1/embeddingsapi.weir.example

Dedicated

Need reserved capacity?

Qwen3 Embedding 8B runs on the shared serverless pool. Dedicated replicas for it are set up on request, with the same API and the same model ID.

Talk to an engineer

Compare

Similar models