Benchmark Almanac

AfroBench-Lite

GPT-5 77.7 leads since August 2025
Not in model cards

Models complete language understanding, question answering, reasoning, and generation tasks across 14 African languages. Seven datasets use task-specific metrics, including accuracy, F1, character-level translation scores, and exact match.

This reduced task and language selection differs from the full 64-language AfroBench. The release month refers to the opened June 2025 paper revision. Per-task metrics must remain separate.

Released
June 2025
Kind
Knowledge in many languages
Level
School
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
Many languages
In and out
Text → Text
Interaction
Single turn
Items from
Another benchmark
Refresh
One fixed set
Run by
Anyone
Based on
AfroBench
In views
Many languages
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
November 2025
Scores refreshed
November 18th, 2026

In the authors’ words

Large-scale multilingual evaluations, such as MEGA, often include only a handful of African languages due to the scarcity of high-quality evaluation data and the limited discoverability of existing African datasets. This lack of representation hinders comprehensive LLM evaluation across a diverse range of languages and tasks. To address these challenges, we introduce AfroBench – a multi-task benchmark for evaluating the performance of LLMs across 64 African languages, 15 tasks and 22 datasets. AfroBench consists of nine natural language understanding datasets, six text generation datasets, six knowledge and question answering tasks, and one mathematical reasoning task. We present results comparing the performance of prompting LLMs to fine-tuned baselines based on BERT and T5-style models. Our results suggest large gaps in performance between high-resource languages, such as English, and African languages across most tasks; but performance also varies based on the availability of monolingual data resources. Our findings confirm that performance on African languages continues to remain a hurdle for current LLMs, underscoring the need for additional efforts to close this gap. https://mcgill-nlp.github.io/AfroBench/

Abstract of AfroBench: How Good are Large Language Models on African Languages? (arXiv:2311.07978, CC0)
24
Entries on its own leaderboard, 17 models we track

GPT-5 leads with 77.7 among models released by November 2025 GPT-5 leads with 77.7 among models released by November 2025

Best score per entry from the benchmark's own leaderboard, read on November 18th, 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. This view shows only models we track, each at its best.

The newest model with a score here came out in November 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-5 77.7
  2. 2 Gemini 3 Pro 76.0
  3. 3 Gemini 2.5 Pro Jun 2025 74.5
  4. 4 Claude Sonnet 4.5 70.6
  5. 5 Claude Sonnet 4 68.7
  6. 6 Gemini 2.5 Flash Sep 2025 66.7
  7. 7 Gemini 2.0 Flash Feb 2025 66.4
  8. 8 GPT-4o (Aug) 65.8
  9. 9 GPT-4.1 65.7
  10. 10 Gemini 1.5 Pro Sep 2024 62.8
  11. 11 Claude 3.7 Sonnet 60.3
  12. 12 Gemma 3 27B 51.0
  13. 13 LLaMa 4 405B 49.7
  14. 14 Gemma 2 27B 44.9
  15. 15 Aya-101 13B 43.4
  16. 16 Gemma 2 9B 40.8
  17. 17 Llama 3.1-70B 40.0
  18. 18 LLaMAX3 8B 28.8
  19. 19 Llama 3.1-8B 25.8
  20. 20 Gemma1.1 7b 23.7
  21. 21 Llama 3-8B 21.4
  22. 22 Lugha-Llama 8B 21.0
  23. 23 Llama 2-7B 16.2
  24. 24 AfroLLaMa 8B 16.1

Source: Official AfroBench-Lite leaderboard

#EntryScoreSettingDateSource
1 GPT-5 77.7 Aug 2025 Official leaderboard
2 Gemini 3 Pro 76.0 Nov 2025 Official leaderboard
3 Gemini 2.5 Pro Jun 2025 74.5 Jun 2025 Official leaderboard
4 Claude Sonnet 4.5 70.6 Sep 2025 Official leaderboard
5 Claude Sonnet 4 68.7 May 2025 Official leaderboard
6 Gemini 2.5 Flash Sep 2025 66.7 Sep 2025 Official leaderboard
7 Gemini 2.0 Flash Feb 2025 66.4 Feb 2025 Official leaderboard
8 GPT-4o (Aug) 65.8 Official leaderboard
9 GPT-4.1 65.7 Apr 2025 Official leaderboard
10 Gemini 1.5 Pro Sep 2024 62.8 Sep 2024 Official leaderboard
11 Claude 3.7 Sonnet 60.3 Feb 2025 Official leaderboard
12 Gemma 3 27B 51.0 Mar 2025 Official leaderboard
13 LLaMa 4 405B 49.7 Official leaderboard
14 Gemma 2 27B 44.9 Jun 2024 Official leaderboard
15 Aya-101 13B 43.4 Official leaderboard
16 Gemma 2 9B 40.8 Jun 2024 Official leaderboard
17 Llama 3.1-70B 40.0 Jul 2024 Official leaderboard
18 LLaMAX3 8B 28.8 Official leaderboard
19 Llama 3.1-8B 25.8 Jul 2024 Official leaderboard
20 Gemma1.1 7b 23.7 Official leaderboard
21 Llama 3-8B 21.4 Apr 2024 Official leaderboard
22 Lugha-Llama 8B 21.0 Official leaderboard
23 Llama 2-7B 16.2 Jul 2023 Official leaderboard
24 AfroLLaMa 8B 16.1 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 77.7 Aug 2025 Official leaderboard
2 Gemini 3 Pro 76.0 Nov 2025 Official leaderboard
3 Gemini 2.5 Pro Jun 2025 74.5 Jun 2025 Official leaderboard
4 Claude Sonnet 4.5 70.6 Sep 2025 Official leaderboard
5 Claude Sonnet 4 68.7 May 2025 Official leaderboard
6 Gemini 2.5 Flash Sep 2025 66.7 Sep 2025 Official leaderboard
7 Gemini 2.0 Flash Feb 2025 66.4 Feb 2025 Official leaderboard
8 GPT-4.1 65.7 Apr 2025 Official leaderboard
9 Gemini 1.5 Pro Sep 2024 62.8 Sep 2024 Official leaderboard
10 Claude 3.7 Sonnet 60.3 Feb 2025 Official leaderboard
11 Gemma 3 27B 51.0 Mar 2025 Official leaderboard
12 Gemma 2 27B 44.9 Jun 2024 Official leaderboard
13 Gemma 2 9B 40.8 Jun 2024 Official leaderboard
14 Llama 3.1-70B 40.0 Jul 2024 Official leaderboard
15 Llama 3.1-8B 25.8 Jul 2024 Official leaderboard
16 Llama 3-8B 21.4 Apr 2024 Official leaderboard
17 Llama 2-7B 16.2 Jul 2023 Official leaderboard

Rankings by part

We read 7 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Intent, MMLU and Math. GPT-5 leads 4 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Intent

17 models we track have a score here.

  1. 1 Gemini 3 Pro ↑1 88.8
  2. 2 Gemini 2.5 Pro Jun 2025 ↑1 88.0
  3. 3 Gemini 2.5 Flash Sep 2025 ↑3 87.4
  4. 4 GPT-5 ↓3 87.3
  5. 5 GPT-4.1 ↑3 82.9
  6. 6 Claude Sonnet 4.5 ↓2 79.3
  7. 7 Claude Sonnet 4 ↓2 77.9
  8. 8 Gemini 1.5 Pro Sep 2024 ↑1 76.3
  9. 9 Gemini 2.0 Flash Feb 2025 ↓2 74.4
  10. 10 Claude 3.7 Sonnet 72.2

3 models have held number one on the official leaderboard since its release in June 2025

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic Google DeepMind OpenAI
Own leaderboard Not tested since Nov 2025
GPT-5 77.7

Source: Official AfroBench-Lite leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard
0.00 20.0 40.0 60.0 80.0 2024 2025 2026 0.00 20.0 40.0 60.0 80.0 2024 2025 2026

Source: Official AfroBench-Lite 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 12 months ago, in November 2025, for Gemini 3 Pro. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (24 entries).
  • Pass Models scored 17 models we track, plus 7 other entries on its leaderboard.
  • Warning Newest score The newest came 12 months ago, in November 2025, for Gemini 3 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on huggingface.co: 17 rows for models we track and 7 other entries at the last read.
  • Pass Last read November 18th, 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