# MKQA-11 multilingual retrieval (MKQA-11)

> A display-only multilingual QA retrieval benchmark reported by Liquid AI for LFM2.5 retriever models, using Recall@20 across 11 languages.

Canonical page: https://benchlm.ai/benchmarks/mkqa11

- Category: [Multilingual](/multilingual)
- Last updated: September 27, 2026

## About MKQA-11

- Year: 2026
- Tasks: Cross-lingual open-domain QA retrieval
- Format: Recall@20 average
- Difficulty: Multilingual retrieval
- Paper: [LFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search](https://www.liquid.ai/blog/lfm2-5-retrievers)

Liquid reports MKQA-11 average Recall@20 across Arabic, German, English, Spanish, French, Italian, Japanese, Korean, Norwegian, Portuguese, and Swedish. BenchLM stores the average as a display-only retrieval signal.

MKQA-11 is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (2 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [LFM2.5-ColBERT-350M](/models/lfm2-5-colbert-350m) | LiquidAI | 69.4% |
| 2 | [LFM2.5-Embedding-350M](/models/lfm2-5-embedding-350m) | LiquidAI | 69.1% |

## FAQ

### What does MKQA-11 measure?

A display-only multilingual QA retrieval benchmark reported by Liquid AI for LFM2.5 retriever models, using Recall@20 across 11 languages.

### Which model scores highest on MKQA-11?

LFM2.5-ColBERT-350M by LiquidAI currently leads with a score of 69.4% on MKQA-11.

### How many models are evaluated on MKQA-11?

2 AI models have been evaluated on MKQA-11 on BenchLM.

### Does MKQA-11 affect BenchLM's overall score?

Not directly. MKQA-11 is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Compare Top Models on MKQA-11

- [LFM2.5-ColBERT-350M vs LFM2.5-Embedding-350M](/compare/lfm2-5-colbert-350m-vs-lfm2-5-embedding-350m)