# The best multilingual embedding model for RAG

> Multilingual RAG means questions and documents in several languages, sometimes different ones. What multilingual retrieval needs, the pitfalls of language bias and tokenization, how to test per language, and a benchmark on a five-language help center.

Published 2026-10-06, updated 2026-11-14 by Truvec team. Canonical: https://truvec.dev/blog/best-multilingual-embedding-model/

Most embedding benchmarks are in English, and most RAG tutorials assume English documents. Many real systems aren't: a help center in five languages, a European company's internal docs, a support bot that answers in the user's language. Multilingual retrieval has its own failure modes, and a model's English score says little about them.

This post explains what a multilingual RAG system needs from its embedding model and how to test it. The last section benchmarks the main multilingual models on a help center in English, French, Spanish, German and Arabic.

## Three multilingual situations

**Same language, several languages.** Each user asks in their language, and the answer exists in that language. The model needs to be good in every language you serve, not just on average.

**Cross-lingual.** The user asks in French, the answer only exists in English. The model must map both languages into the same space, so that a French question lands near an English passage. Not all multilingual models do this well.

**Mixed text.** Real documents mix languages: English product names in French articles, code in Arabic tutorials, quotes in another language. The model must not be thrown off by the switch.

Know which situations you have before you test: the right model for one isn't always the right model for the others.

## Pitfalls

**Language bias.** Many models prefer passages in the same language as the question, even when a passage in another language answers it better. In a mixed corpus, this means a French question retrieves French passages that are only roughly relevant, over the exact answer in English.

**Uneven quality across languages.** A model can be excellent in English and Spanish and weak in Arabic or Thai. Averages hide this: always look at results per language.

**Tokenization cost.** Tokenizers built mostly on English text split other languages into more tokens. The same text can cost noticeably more to embed in Arabic or Hindi than in English, and long texts may hit input limits sooner.

**English-only models.** Models such as BGE Base EN are trained on English. On other languages they return poor results that look plausible at first glance.

## How to test

1. **Build the golden set per language**, in proportion to your real traffic. Generated questions should be in the language users actually ask in.
2. **Add cross-lingual questions** if users may ask in a language the answer isn't written in.
3. **Compare per language** before you look at the overall ranking. A model that wins overall but fails in one language you serve is the wrong choice.
4. **Include an English-only model as a baseline.** It shows how much multilingual support is worth on your data.

## Candidates

Cohere's multilingual models (Embed v4, Embed Multilingual v3), Google's Gemini embeddings, Qwen3 Embedding and OpenAI text-embedding-3-large are the usual shortlist. Voyage's models are worth including as well. Check each provider's documentation for the languages it supports.

The benchmark below runs eight models on a five-language help center, with an English-only model as a baseline. Look at the example questions in each language: averages hide which languages each model struggles with. For what each metric in the tables means, see [RAG evaluation metrics explained](/blog/rag-evaluation-metrics-that-matter/); for why a public leaderboard can't settle this for your documents, see [why MTEB alone isn't enough](/blog/mteb-benchmark-validate-embedding-model-before-switching/).

## Benchmark: Multilingual embedding models on a 5-language help center

Real benchmark generated by Truvec (benchmark #15, run 2026-11-14). Raw data: [benchmark.json](https://truvec.dev/blog/best-multilingual-embedding-model/benchmark.json), [benchmark.csv](https://truvec.dev/blog/best-multilingual-embedding-model/benchmark.csv). Full report with every question: [report.pdf](https://truvec.dev/blog/best-multilingual-embedding-model/report.pdf).

- Corpus: MediaWiki help pages in English, French, Spanish, German and Arabic (https://www.mediawiki.org/wiki/Help:Contents), 165 documents, 2,050 chunks
- Chunking: recursive, 1000 chars, 200 overlap
- Golden set: 100 questions generated by gpt-5.6-luna, human-reviewed
- Retrieval: top 5 by cosine similarity; models ranked by MRR, then recall@k

### Key results

- **OpenAI text-embedding-3-large** ranks first, with an MRR of 0.848 and a hit rate of 97.0% in the top 5.
- It leads the runner-up, Cohere Embed Multilingual v3, by 0.037 MRR.
- Hit rate ranges from 70.0% to 97.0% across the 8 models.
- Cheapest model within 2 points of the winner's hit rate: **Cohere Embed v4**, $0.05 to embed this corpus against $0.06.
- Fastest to embed a query: **OpenAI text-embedding-3-small** (5.7 ms per query, measured in batch from our servers).

### Retrieval quality

| Rank | Model | Provider | Hit rate@5 | Recall@5 | MRR | nDCG@5 | Precision@5 |
| --- | --- | --- | ---: | ---: | ---: | ---: | ---: |
| 1 | OpenAI text-embedding-3-large | openai | 97.0% | 96.5% | 0.848 | 0.876 | 19.6% |
| 2 | Cohere Embed Multilingual v3 | cohere | 93.0% | 92.5% | 0.811 | 0.839 | 18.8% |
| 3 | Cohere Embed v4 | cohere | 95.0% | 94.5% | 0.793 | 0.831 | 19.2% |
| 4 | Gemini Embedding 2 | google | 93.0% | 92.5% | 0.781 | 0.815 | 18.6% |
| 5 | Voyage 4 Large | voyage | 95.0% | 95.0% | 0.780 | 0.823 | 19.2% |
| 6 | Qwen3 Embedding 8B | fireworks | 90.0% | 89.5% | 0.749 | 0.785 | 18.0% |
| 7 | OpenAI text-embedding-3-small | openai | 88.0% | 88.0% | 0.734 | 0.768 | 18.0% |
| 8 | BGE Base EN v1.5 | fireworks | 70.0% | 70.0% | 0.583 | 0.610 | 14.4% |

### Hit rate by position

| Model | Top 1 | Top 2 | Top 3 | Top 4 | Top 5 |
| --- | ---: | ---: | ---: | ---: | ---: |
| OpenAI text-embedding-3-large | 78.0% | 85.0% | 91.0% | 93.0% | 97.0% |
| Cohere Embed Multilingual v3 | 74.0% | 82.0% | 87.0% | 91.0% | 93.0% |
| Cohere Embed v4 | 70.0% | 81.0% | 88.0% | 90.0% | 95.0% |
| Gemini Embedding 2 | 68.0% | 83.0% | 86.0% | 91.0% | 93.0% |
| Voyage 4 Large | 68.0% | 79.0% | 88.0% | 91.0% | 95.0% |
| Qwen3 Embedding 8B | 65.0% | 79.0% | 83.0% | 86.0% | 90.0% |
| OpenAI text-embedding-3-small | 64.0% | 76.0% | 83.0% | 84.0% | 88.0% |
| BGE Base EN v1.5 | 52.0% | 59.0% | 62.0% | 66.0% | 70.0% |

### Cost, speed and size

| Model | Price / 1M tokens | Embed the corpus | 1,000 queries | Query embedding | Search p95 | Vectors |
| --- | ---: | ---: | ---: | ---: | ---: | ---: |
| OpenAI text-embedding-3-large | $0.13 | $0.06 | $0.0039 | 8.1 ms | 54 ms | 24.0 MB |
| Cohere Embed Multilingual v3 | $0.10 | $0.04 | $0.0030 | 7.8 ms | 49 ms | 8.0 MB |
| Cohere Embed v4 | $0.12 | $0.05 | $0.0036 | 11 ms | 31 ms | 12.0 MB |
| Gemini Embedding 2 | $0.20 | $0.09 | $0.0060 | 22 ms | 65 ms | 24.0 MB |
| Voyage 4 Large | $0.12 | $0.05 | $0.0036 | 17 ms | 33 ms | 8.0 MB |
| Qwen3 Embedding 8B | $0.10 | $0.04 | $0.0030 | 21 ms | 56 ms | 32.0 MB |
| OpenAI text-embedding-3-small | $0.02 | $0.0087 | $0.0006 | 5.7 ms | 52 ms | 12.0 MB |
| BGE Base EN v1.5 | $0.0080 | $0.0032 | $0.0002 | 18 ms | 25 ms | 6.0 MB |

### Score separation

| Model | Relevant chunk | Best distractor | Gap |
| --- | ---: | ---: | ---: |
| OpenAI text-embedding-3-large | 0.660 | 0.621 | 0.039 |
| Cohere Embed Multilingual v3 | 0.730 | 0.696 | 0.035 |
| Cohere Embed v4 | 0.628 | 0.599 | 0.027 |
| Gemini Embedding 2 | 0.813 | 0.800 | 0.012 |
| Voyage 4 Large | 0.616 | 0.602 | 0.013 |
| Qwen3 Embedding 8B | 0.805 | 0.784 | 0.019 |
| OpenAI text-embedding-3-small | 0.693 | 0.648 | 0.038 |
| BGE Base EN v1.5 | 0.761 | 0.728 | 0.033 |

### Example questions

Rank of the first relevant chunk per model ("miss": not in the top 5).

| Question | OpenAI text-embedding-3-large | Cohere Embed Multilingual v3 | Cohere Embed v4 | Gemini Embedding 2 | Voyage 4 Large | Qwen3 Embedding 8B | OpenAI text-embedding-3-small | BGE Base EN v1.5 |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
| ¿Cómo decide WebFonts si usa una fuente de mi dispositivo, la descarga o la recupera de la caché? | 1 | 1 | 1 | 4 | 3 | 5 | 2 | miss |
| ¿Qué permiso se necesita para crear listas de envío de MassMessage, y dónde se puede crear una lista nueva? | 5 | miss | 4 | 3 | 3 | 2 | 3 | miss |
| Quelles sont les étapes pour commencer une nouvelle traduction avec l'outil de traduction de contenu ? | 3 | 3 | miss | 2 | 1 | 5 | 5 | miss |
| كيف أشترك في موضوع نقاش ثم أوقف تلقي الإشعارات عنه لاحقًا؟ | 4 | 2 | 2 | miss | 5 | 4 | 3 | miss |
| Wie setze ich mein Wiki-Passwort mit meinem Benutzernamen oder meiner E-Mail-Adresse zurück? | 1 | 3 | 3 | 1 | 2 | 5 | 2 | miss |
| What are the main navigation areas available on a MediaWiki page? | 3 | miss | 5 | 3 | 1 | miss | miss | 2 |
| ¿Durante cuánto tiempo se conserva la dirección IP asociada a una edición hecha con una cuenta temporal? | 3 | 1 | 1 | 1 | 1 | 5 | 2 | 4 |
| ما المعلومات التي تتلقاها خدمات الترجمة الخارجية، وما القيد المتعلق بالبيانات الشخصية عند ربطها؟ | 1 | 2 | 3 | 4 | 4 | 3 | miss | miss |

### Notes

- The corpus is 33 help pages from mediawiki.org (Help namespace, public domain under CC0), each taken in English, French, Spanish, German and Arabic: 165 documents, fetched as plain text on 14 November 2026. Pages were picked at random among the 106 help pages translated into all five languages, keeping those between 2,500 and 40,000 characters whose Arabic version is mostly in Arabic.
- The 100 questions cover the five languages: 33 in English, 21 in French, 16 in German, 15 in Spanish and 15 in Arabic. The generator wrote every question in English, so during review we translated 67 of them into the language of the passage they were written from.
- Each question expects the passage in its own language. The same passage exists in the four other languages: a model that returns one of those instead gets no credit for it, even though it answers the question. Scores here measure same-language retrieval inside a mixed-language corpus, and are lower than they would be on five separate corpora.
- BGE Base EN v1.5 is English-only: it is included as a baseline to show what a monolingual model loses. It reads at most 380 tokens per chunk in Truvec and Cohere Embed Multilingual v3 at most 512, so both saw some of the longer chunks cut short.
- With 100 questions, a gap of a couple of points between two models is within the noise, and each language has only 15 to 33 questions: per-language differences need more questions than this.

### Metric definitions

- **Hit rate**: Share of questions with at least one relevant chunk in the top k results. The most direct answer to "does the right passage reach the LLM?".
- **Recall**: Share of the relevant chunks found in the top k, averaged over questions. Matters when an answer spans several chunks.
- **MRR**: Mean reciprocal rank: 1 when the first relevant chunk is ranked first, 1/2 when second, 0 when missing. Rewards putting the answer at the top.
- **nDCG**: Normalized discounted cumulative gain: how close the ranking of all relevant chunks is to the ideal ranking.
- **Precision**: Share of the top k results that are relevant. Low precision means more noise sent to the LLM.
