# MKQA-11 leaderboard

Multilingual benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/leaderboards/benchmarks/multilingual-mkqa11

MKQA-11 multilingual retrieval. Every model the catalog carries a published MKQA-11 value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Multilingual |
| Measure | Recall@20 average |
| Tasks | Cross-lingual open-domain QA retrieval |
| Difficulty | Multilingual retrieval |
| Published by | [LFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search](https://www.liquid.ai/blog/lfm2-5-retrievers) |

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

## MKQA-11 ranking

2 models with a published MKQA-11 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Recall@20 average | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [LFM2.5-ColBERT-350M](https://modelscale.dev/models/lfm2-5-colbert-350m) | LiquidAI | 69.4 | Observed 2026-11-21 · source benchlm:models |
| 2 | [LFM2.5-Embedding-350M](https://modelscale.dev/models/lfm2-5-embedding-350m) | LiquidAI | 69.1 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
