Benchmark Almanac

NaijaSenti

McGill-NLP/AfriE5-Large-instruct 49.1% tops its leaderboard Qwen3-Embedding-8B 48.8% leads the models we track
Not in model cards

Models classify the sentiment of posts written in Hausa, Igbo, Nigerian Pidgin, and Yoruba.

MTEB reports accuracy on its NaijaSenti task.

Released
January 2022
Kind
Reading in many languages
Level
Everyday
Scored
Checked automatically
Answers
Prediction
Tasks
All public
Measures
Capability
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Last model added
June 2025
Scores refreshed
November 20th, 2026
20
Entries on MTEB’s leaderboard, 1 model we track

McGill-NLP/AfriE5-Large-instruct tops the leaderboard with 49.1% Qwen3-Embedding-8B leads with 48.8%

Best score per entry from MTEB’s leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Qwen3-Embedding-8B, at 48.8%. This view shows only models we track, each at its best.

  1. 1 McGill-NLP/AfriE5-Large-instruct 49.1%
  2. 2 Qwen3-Embedding-8B 48.8%
  3. 3 intfloat/multilingual-e5-large-instruct 47%
  4. 4 GritLM/GritLM-7B 46.9%
  5. 5 intfloat/e5-mistral-7b-instruct 46.8%
  6. 6 microsoft/harrier-oss-v1-0.6b 46.2%
  7. 7 intfloat/multilingual-e5-large 45.9%
  8. 8 BAAI/bge-m3 45%
  9. 9 intfloat/multilingual-e5-base 43.9%
  10. 10 sentence-transformers/LaBSE 43.8%
  11. 11 codefuse-ai/F2LLM-v2-8B 43.1%
  12. 12 intfloat/multilingual-e5-small 42.9%
  13. 13 sentence-transformers/all-MiniLM-L6-v2 40.5%
  14. 14 sentence-transformers/all-mpnet-base-v2 40.3%
  15. 15 minishlab/potion-multilingual-128M 40.1%
  16. 16 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 40%
  17. 17 sentence-transformers/all-MiniLM-L12-v2 39.9%
  18. 18 sentence-transformers/static-similarity-mrl-multilingual-v1 39.8%
  19. 19 codefuse-ai/F2LLM-v2-80M 38.5%
  20. 20 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 38.1%

Source: MTEB’s leaderboard

Source: MTEB’s leaderboard

#EntryScoreSettingDateLinkSource
1 McGill-NLP/AfriE5-Large-instruct open weights, 0.6B 49.1% huggingface.co Official leaderboard
2 Qwen3-Embedding-8B 48.8% Jun 2025 huggingface.co Official leaderboard
3 intfloat/multilingual-e5-large-instruct open weights, 0.6B 47% huggingface.co Official leaderboard
4 GritLM/GritLM-7B open weights, 7.2B 46.9% huggingface.co Official leaderboard
5 intfloat/e5-mistral-7b-instruct open weights, 7.1B 46.8% huggingface.co Official leaderboard
6 microsoft/harrier-oss-v1-0.6b open weights, 0.6B 46.2% huggingface.co Official leaderboard
7 intfloat/multilingual-e5-large open weights, 0.6B 45.9% huggingface.co Official leaderboard
8 BAAI/bge-m3 open weights, 0.6B 45% huggingface.co Official leaderboard
9 intfloat/multilingual-e5-base open weights, 0.3B 43.9% huggingface.co Official leaderboard
10 sentence-transformers/LaBSE open weights, 0.5B 43.8% huggingface.co Official leaderboard
11 codefuse-ai/F2LLM-v2-8B open weights, 7.6B 43.1% huggingface.co Official leaderboard
12 intfloat/multilingual-e5-small open weights, 0.1B 42.9% huggingface.co Official leaderboard
13 sentence-transformers/all-MiniLM-L6-v2 open weights, 0B 40.5% huggingface.co Official leaderboard
14 sentence-transformers/all-mpnet-base-v2 open weights, 0.1B 40.3% huggingface.co Official leaderboard
15 minishlab/potion-multilingual-128M open weights, 0.1B 40.1% huggingface.co Official leaderboard
16 sentence-transformers/paraphrase-multilingual-mpnet-base-v2 open weights, 0.3B 40% huggingface.co Official leaderboard
17 sentence-transformers/all-MiniLM-L12-v2 open weights, 0B 39.9% huggingface.co Official leaderboard
18 sentence-transformers/static-similarity-mrl-multilingual-v1 open weights, 0.1B 39.8% huggingface.co Official leaderboard
19 codefuse-ai/F2LLM-v2-80M open weights, 0.1B 38.5% huggingface.co Official leaderboard
20 sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2 open weights, 0.1B 38.1% huggingface.co Official leaderboard
#ModelScoreSettingDateLinkSource
1 Qwen3-Embedding-8B 48.8% Jun 2025 huggingface.co Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On MTEB’s leaderboard
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Qwen3-Embedding-8B 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Qwen3-Embedding-8B

Source: MTEB’s leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Multilingual

All 545 in Multilingual
  • MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
  • WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET. K-EXAONE 90.5%
  • MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language. Qwen3.7-Max 87%
  • CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories. Solar Open 2 90.7%
  • Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
  • MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 17 months ago, in June 2025, for Qwen3-Embedding-8B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (20 entries).
  • Pass Models scored 1 model we track, plus 19 other entries on its leaderboard.
  • Warning Newest score The newest came 17 months ago, in June 2025, for Qwen3-Embedding-8B.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 48.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on MTEB: 1 row for models we track and 19 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check