AI Model Watch
Models / Alibaba

Qwen Text-Embedding v4

GA

Alibaba Model Studio (DashScope) commercial text/code embedding model, based on the Qwen3-Embedding family. Output is an embedding vector, not tokens, so there is no output-token price. Max input 8,192 tokens per text; batch up to 10 texts per call. Flexible (MRL) output dimensions: 2048 / 1536 / 1024 (default) / 768 / 512 / 256 / 128 / 64. Supports 100+ languages incl. programming languages, plus task instructions and sparse vectors. The $0.07/1M figure is the Singapore/Hong Kong international-region rate; Beijing region is $0.072/1M. The open-weight Qwen3-Embedding checkpoints (0.6B/4B/8B, Apache-2.0 on Hugging Face) are separate downloadable models, not this hosted API; open_weight here reflects the commercial API model. Docs fetched 2026-07-24.

Qwen Text-Embedding v4 by Alibaba is tracked live in our LLM catalog, with a 8K (8192-token) context window. It is generally available (GA).

Last verified: 13 Aug 2026 · sourced from official provider documentation

Provider
Alibaba
Status
GA
Input price
$0.07 / 1M tokens
Output price
—
Cached input
—
Blended price
—
Context window
8192 tokens (8K)
Max output
—
Embedding dimensions
1024-dim output vectors
Modality
text
Open weights
No — API only
Knowledge cutoff
—
Released
—
API string
text-embedding-v4

Source: Alibaba official documentation ↗

See also

Explore more

All Alibaba models→

Prices, context windows and lifecycle status for every Alibaba model we track.

Compare models head-to-head→

Side-by-side price, context and status for the models developers weigh against each other.

FAQ

Qwen Text-Embedding v4 — questions & answers

What is the context window of Qwen Text-Embedding v4?

Qwen Text-Embedding v4 has a 8192-token context window (8K).

Is Qwen Text-Embedding v4 deprecated?

No — Qwen Text-Embedding v4 is generally available (GA) and not currently scheduled for deprecation or retirement in our tracker.

Does Qwen Text-Embedding v4 support image or vision input?

Our catalog lists Qwen Text-Embedding v4's modalities as text; image/vision input is not among them.

Can I self-host Qwen Text-Embedding v4?

No — Qwen Text-Embedding v4 is proprietary. Its weights are not publicly released, so it is available only through Alibaba's API (or hosted partners), not for self-hosting.

What is the API model string for Qwen Text-Embedding v4?

The API model identifier for Qwen Text-Embedding v4 is "text-embedding-v4" when calling Alibaba's API.