Benchmark Almanac

IndicIFEval

GPT-5 82.4% leads since February 2026
Not in model cards

Models follow instructions in 14 Indic languages. Rule-based checks test constraints such as required words, paragraph structure and word counts.

The February 2026 paper compares Gemini, GPT, Aya, Qwen and other models. Gemini 3 Pro led on translated prompts with 90.0% average loose accuracy.

Made by
AI4Bharat
Released
February 2026
Kind
Chat and instructions in many languages
Level
Everyday
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
Many languages
In and out
Text → Text
Interaction
Single turn
Items from
Translated
Refresh
One fixed set
Run by
Anyone
Based on
IFEval
In views
Many languages
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
December 2025
Scores refreshed
November 20th, 2026
19
Entries in its paper, 16 models we track

GPT-5 leads with 82.4% among models released by December 2025 GPT-5 leads with 82.4% among models released by December 2025

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper gives no combined score. The headline is the plain mean of translated and grounded prompt accuracy. 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 December 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 82.4%
  2. 2 GPT-5 mini 77.8%
  3. 3 Gemini 3 Pro 76.3%
  4. 4 Gemini 3 Flash 74.8%
  5. 5 Qwen3-14B 70.5%
  6. 6 Gemma 3 27B 69.7%
  7. 7 Qwen3-32B 69.4%
  8. 8 Gemma 3 12B 67.8%
  9. 9 Qwen3-8B 64.7%
  10. 10 Gemma 3 4B 60.1%
  11. 11 Qwen3-4B 58.1%
  12. 12 meta-llama/Llama-3.1-8B-Instruct 51.1%
  13. 13 Qwen3-1.7B 45.2%
  14. 14 Aya Expanse 32B 44.3%
  15. 15 meta-llama/Llama-3.2-3B-Instruct 44.1%
  16. 16 Gemma 3 1B 37.6%
  17. 17 Qwen3-0.6B 37%
  18. 18 meta-llama/Llama-3.2-1B-Instruct 35.5%
  19. 19 Aya Expanse 8B 32.9%

Source: The IndicIFEval paper

#EntryScoreSettingDateSource
1 GPT-5 82.4% Aug 2025 Official leaderboard
2 GPT-5 mini 77.8% Aug 2025 Official leaderboard
3 Gemini 3 Pro 76.3% Nov 2025 Official leaderboard
4 Gemini 3 Flash 74.8% Dec 2025 Official leaderboard
5 Qwen3-14B 70.5% Apr 2025 Official leaderboard
6 Gemma 3 27B 69.7% Mar 2025 Official leaderboard
7 Qwen3-32B 69.4% Apr 2025 Official leaderboard
8 Gemma 3 12B 67.8% Mar 2025 Official leaderboard
9 Qwen3-8B 64.7% Apr 2025 Official leaderboard
10 Gemma 3 4B 60.1% Mar 2025 Official leaderboard
11 Qwen3-4B 58.1% Apr 2025 Official leaderboard
12 meta-llama/Llama-3.1-8B-Instruct 51.1% Official leaderboard
13 Qwen3-1.7B 45.2% Apr 2025 Official leaderboard
14 Aya Expanse 32B 44.3% Oct 2024 Official leaderboard
15 meta-llama/Llama-3.2-3B-Instruct 44.1% Official leaderboard
16 Gemma 3 1B 37.6% Mar 2025 Official leaderboard
17 Qwen3-0.6B 37% Apr 2025 Official leaderboard
18 meta-llama/Llama-3.2-1B-Instruct 35.5% Official leaderboard
19 Aya Expanse 8B 32.9% Oct 2024 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 82.4% Aug 2025 Official leaderboard
2 GPT-5 mini 77.8% Aug 2025 Official leaderboard
3 Gemini 3 Pro 76.3% Nov 2025 Official leaderboard
4 Gemini 3 Flash 74.8% Dec 2025 Official leaderboard
5 Qwen3-14B 70.5% Apr 2025 Official leaderboard
6 Gemma 3 27B 69.7% Mar 2025 Official leaderboard
7 Qwen3-32B 69.4% Apr 2025 Official leaderboard
8 Gemma 3 12B 67.8% Mar 2025 Official leaderboard
9 Qwen3-8B 64.7% Apr 2025 Official leaderboard
10 Gemma 3 4B 60.1% Mar 2025 Official leaderboard
11 Qwen3-4B 58.1% Apr 2025 Official leaderboard
12 Qwen3-1.7B 45.2% Apr 2025 Official leaderboard
13 Aya Expanse 32B 44.3% Oct 2024 Official leaderboard
14 Gemma 3 1B 37.6% Mar 2025 Official leaderboard
15 Qwen3-0.6B 37% Apr 2025 Official leaderboard
16 Aya Expanse 8B 32.9% Oct 2024 Official leaderboard

Rankings by part

We read 31 parts from the benchmark’s paper, each ranking models on one part of the test, such as IndicIFEval-Trans, IndicIFEval-Trans as and IndicIFEval-Trans bn. Gemini 3 Pro leads 11 of them, and 4 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

IndicIFEval-Trans

16 models we track have a score here.

  1. 1 Gemini 3 Pro ↑2 90%
  2. 2 GPT-5 ↓1 88.9%
  3. 3 Gemini 3 Flash ↑1 87.5%
  4. 4 GPT-5 mini ↓2 79.8%
  5. 5 Gemma 3 27B ↑1 76.1%
  6. 6 Gemma 3 12B ↑2 73.6%
  7. 7 Qwen3-32B 72.8%
  8. 8 Qwen3-14B ↓3 71.1%
  9. 9 Qwen3-8B 62.7%
  10. 10 Gemma 3 4B 60.1%

Number one over time

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

OpenAI
Benchmark paper Not tested since Feb 2026
GPT-5 82.4%

Source: The IndicIFEval paper

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
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5 Qwen3-14B Aya Expanse 32B 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GPT-5 Qwen3-14B

Source: Official IndicIFEval 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 11 months ago, in December 2025, for Gemini 3 Flash. 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 benchmark’s paper (19 entries).
  • Pass Models scored 16 models we track, plus 3 other entries on its leaderboard.
  • Warning Newest score The newest came 11 months ago, in December 2025, for Gemini 3 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Pass Headroom The best is 82.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 16 rows for models we track and 3 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