Back to models

Qwen3 Embedding 4B

Available Serverless
DOS.AI

Self-hosted multilingual text embedding (2560 dims), strong on Vietnamese. Priced at our OpenRouter fallback cost.

Modalities
Input / Output Price
$0.02/$0.00/ 1M
Context Length
33K tokens
Parameters
-

Pricing

Transparent pass-through token pricing per million tokens. DOS.AI strictly enforces a Zero-Markup retail policy.

Input / Prompt
$0.02/ 1M tokens
Published official list price
Output / Completion
$0.00/ 1M tokens
Zero-markup retail rate
Prompt Cache Read
$0.002/ 1M tokens
Up to 90% cache discount
Context Window
33Ktokens
Full context window supported

Uptime & Reliability

Live upstream health and gateway reliability. Automatic multi-provider failover protects your production traffic.

All Endpoints Operational
SLA Target: 99.99%Last 30 Days: 100.0%
30 days ago100.0% uptime over past monthToday

Benchmarks & Evaluations

Standardized capability benchmarks and live inference speed telemetry across key engineering tasks.

Evaluation DimensionPerformance TierRating / ScoreStatus
Reasoning & Logic (GPQA Diamond / MMLU-Pro)
Frontier scientific problem solving & complex multi-step reasoning
High CompetencyStandard High (74% - 82%)Verified
Agentic Code Generation (SWE-bench / HumanEval)
Full-repository bug fixing & autonomous code synthesis
Standard HighSupported (85%+ HumanEval)Verified
Multimodal Vision Understanding (MMMU / MathVista)
College-level multi-discipline visual perception & diagram reasoning
Text SpecializationText OnlyN/A
Tool Calling & Instruction Following (Exacto / BFCL)
Berkeley Function Calling Leaderboard & strict JSON schema validation
Native JSON & Tools100% Strict Schema ValidatedVerified

API Integration

Use the DOS API to integrate Qwen3 Embedding 4B into your applications. Compatible with standard OpenAI SDKs for easy drop-in migration.

</>Run inference
curl https://api.dos.ai/v1/embeddings \
  -H "Authorization: Bearer $DOS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-embedding-4b",
    "input": ["Text to embed"],
    "encoding_format": "float"
  }'