Embeddings multilingual-e5-large für ein deutsches RAG
multilingual-e5-large für ein RAG auf Deutsch: 1024 Dimensionen, 512 Tokens, Präfixe query:/passage: erforderlich, Kosinus-Distanz. MIT-Lizenz.
| Model | intfloat/multilingual-e5-large |
|---|---|
| Dimensions | 1024 |
| Max tokens | 512 |
| Distance | cosine |
| Languages | de, en, fr, es, it, nl |
| Prefixes | query: / passage: |
| Chunk size | 350 |
Notes
Die Präfixe „query: “ und „passage: “ sind Pflicht: ohne sie wird die Suche deutlich schlechter. Höchstens 512 Tokens pro Text: Dokumente in Abschnitte von etwa 350 Wörtern teilen. Vektoren normalisieren und Kosinus-Distanz verwenden. Von Microsoft (intfloat) unter MIT-Lizenz veröffentlicht.
Estimated cost
Input tokens per call : ≈ 249
| Model | Per call | Per 1,000 calls |
|---|---|---|
| Mistral Large · Mistral AI | €0.00011 | €0.11 |
| DeepSeek Flash · DeepSeek | €0.00007 | €0.07 |
| DeepSeek V4 Pro · DeepSeek | €0.00029 | €0.29 |
| GPT-6 Luna · OpenAI | €0.00002 | €0.02 |
| GPT-6 Sol · OpenAI | €0.00044 | €0.44 |
| Claude Sonnet 5.5 · Anthropic | €0.00044 | €0.44 |
| Claude Opus 5.5 · Anthropic | €0.00088 | €0.88 |
| Self-hosted open-source model (Ollama, vLLM) | €0 in API fees (server cost only) | |
Input cost only, excluding the model's answer. Providers' public standard prices (Mistral AI, DeepSeek, OpenAI, Anthropic) converted from USD to euros at the ECB rate of September 29, 2026. Estimate: 1 token ≈ 3.6 characters.
Common to all 3 regions
Sources under a compatible licence.
Sources : Modell intfloat/multilingual-e5-large, MIT-Lizenz (Hugging-Face-Modellkarte)
Commercial use allowed · credit the author · changes allowed.
Text or data only: no access to files, network or commands.
Hosted in France (Scaleway, Paris). No transfer outside the EU.
European Union · 5/5
No personal data.
No training data: no summary required.
United States · 5/5
No personal data.
No training data: no documentation to publish.
China · 5/5
No personal data.
No training data; AI-generated content published in China must be labelled (2025).
Indicative summary as of 30/09/2026, not a legal certification. Method and sources →
Statistics
Likes : 0 · Comments : 0
Click a counter to show or hide it, hover the chart for details. One view per visitor per day, bots excluded; counting started on 30 September 2026.
- Identifier
- contextetech--multilingual-e5-large-rag-deutsch
- Type
- Embeddings
- File
- multilingual-e5-large-rag-deutsch.embedding.json
- Format
- JSON (application/json)
- Size
- 683 bytes
- Encoding
- UTF-8
- Estimated tokens
- ≈ 189
- Language
- German
- License
- MIT
- Commercial use
- Allowed
- Personal data
- None
- Version
- 1.0
- Published on
- October 3, 2026 at 5:20 AM
- Updated on
- October 3, 2026 at 5:20 AM
- Storage
- in database
- Uses
- 0
- Likes
- 0
multilingual-e5-large-rag-deutsch.embedding.json
{
"format": "contextetech/embedding/v1",
"id": "contextetech/multilingual-e5-large-rag-deutsch",
"description": "multilingual-e5-large für ein RAG auf Deutsch: 1024 Dimensionen, 512 Tokens, Präfixe query:/passage: erforderlich, Kosinus-Distanz. MIT-Lizenz.",
"tags": [
"rag",
"mehrsprachig",
"semantische-suche"
],
"license": "MIT",
"language": "de",
"model": "intfloat/multilingual-e5-large",
"dimensions": 1024,
"maxTokens": 512,
"distance": "cosine",
"languages": [
"de",
"en",
"fr",
"es",
"it",
"nl"
],
"prefixes": {
"query": "query: ",
"passage": "passage: "
},
"chunkSize": 350,
"benchmarks": []
}Embed (Python)
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("intfloat/multilingual-e5-large")
q = model.encode(["query: your input here"], normalize_embeddings=True)
docs = model.encode(["passage: …"], normalize_embeddings=True)
print(q.shape) # (1, 1024)
print(q @ docs.T) # similarité
Community
No comments yet. Share your feedback, it will help the next person.