Alibaba Cloud: Qwen3.7 Text Embedding Flash

Qwen3.7 Text Embedding Flash offers a lighter vector representation for multilingual search. Its smaller dimension range is useful for large collections where embedding throughput and index size are practical constraints.

Compare models

qwen3.7-text-embedding-flash

AvailableAlibaba CloudEmbeddingSync

Input / Output: $0.0184 / $0.00
Context: 128K
Modalities: Embedding

About Qwen3.7 Text Embedding Flash

Qwen3.7 Text Embedding Flash is the lightweight version of Alibaba's Qwen3.7 text embedding model, built for large collections where throughput and index size matter. Alibaba says it supports task instructions and lets you choose vector dimensions from 256 to 1024, a narrower range than the full model's 256 to 2560. It targets multilingual search at a lower footprint.

Where it works well

Smaller maximum vector size keeps indexes compact for collections with many millions of items.
Supports task instructions, so queries can describe the retrieval goal.
Alibaba positions it for throughput-sensitive workloads such as bulk embedding.
Covers the same multilingual text use as the full Qwen3.7 model.

When to choose another model

Alibaba describes it as the lighter model; the full Qwen3.7 Text Embedding is its highest-performing one.
Its dimensions top out at 1024, and sparse vector output is listed for the full model.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

EmbeddingTokenLab endpoint

POST/v1/embeddings

curl -X POST "https://api.tokenlab.sh/v1/embeddings" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "qwen3.7-text-embedding-flash",
  "input": "The quick brown fox jumps over the lazy dog."
}'

Parameters

Embedding

Parameter	Values and limits	Default	Notes
`dimensions`	256, 512, 768, 1024	Not provided	Not provided
`encoding_format`	float	Not provided	Not provided
`input`	Inputs: ≥ 1, ≤ 20	Not provided	Each input row is limited to 128000 tokens.
`user`	Not provided	Not provided	Not provided

TokenLab exposes this model through the OpenAI-compatible /v1/embeddings route.

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Pricing

per 1M tokens

Verified price: —
Official price: Input$0.0184per 1M tokensOutput$0.00per 1M tokens

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Input	$0.0184	—	—
Output	$0.00	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Nov 13, 2026, 4:48 AM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Qwen3.7 Text Embedding Flash in Console with a prompt ready to edit or send.

Help me try qwen3.7-text-embedding-flash with text-embedding at /v1/embeddings. Show the result and cost.

Use cases

Bulk indexing
Embed a very large archive of documents quickly and store compact vectors.
Product search
Power multilingual search over a big catalogue where query volume is high.
Deduplication
Cluster near-identical text across a large corpus using modest vectors, so duplicates are found cheaply.

Prompt examples

Embed these 500,000 product descriptions at 512 dimensions for catalogue search.

Embed the query 'waterproof hiking jacket' with an instruction for e-commerce retrieval.

Embed this set of news headlines at 256 dimensions for duplicate detection.

Cost calculator

Monthly input tokens1M tokens

Monthly output tokens0.5M tokens

Estimated monthly cost$0.018382

FAQ

What is the difference from the full Qwen3.7 model?

Flash is the lighter, cost- and throughput-oriented version, with dimensions from 256 to 1024. The full model reaches 2560 dimensions and is Alibaba's top-performing embedding model.

What dimensions can I choose?

Alibaba says any size from 256 to 1024 is available. Smaller vectors cut storage and speed up similarity search, while larger ones keep more detail.

Does it support instructions?

Yes. Alibaba lists task instruction support, so you can state what the retrieval is for, such as a product search or a help-centre lookup. Write the instruction in plain language.

Is Flash enough for production search?

Often yes for large collections, but verify on your own queries. Measure recall at your chosen dimension and compare it with the full model before committing.

How much does Qwen3.7 Text Embedding Flash cost?

On TokenLab, Qwen3.7 Text Embedding Flash costs Input $0.0184 / Output $0.00 per 1M tokens. The pricing table above shows the full breakdown.

Which endpoint should Qwen3.7 Text Embedding Flash use?

Use https://api.tokenlab.sh/v1/embeddings for Qwen3.7 Text Embedding Flash. The request example below shows the matching code shape.

Which operations does Qwen3.7 Text Embedding Flash support?

Qwen3.7 Text Embedding Flash supports Embedding. Select an operation above to see its endpoint and request example.