whatmodel.sh / Flan-T5 XXL

Flan-T5 XXL

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Flan-T5 XXL is an open-weights language model from Google, released October 2022. Its published size is 11.3B parameters. Licence: apache-2.0.

Catalogue identity notes

Instruction-finetuned T5 v1.1 LM-adapted checkpoint released with Chung et al. 2022.

33.3±1.3#615 of 719 LLMsfrom 8 benchmark results across 4 sources

$ how it compares

The LLMs ranked either side of Flan-T5 XXL, out of 828 with a score, then more from Google.

#modelscorewhereRAMreleased
709Qwen3 0.6B Alibaba33±1local~1.8GB2025-04
710Llama 2 13B Chat Meta33±1local~9.8GB2023-07
711Flan-T5 XXL Google33±1local~49GB2022-10
712StarCoder2 3B BigCode33±1local~3.4GB2024-02
713Granite 3B Code Base (2K) IBM33±1local~3.7GB2024-04
more from Google
9Gemini 4 Argon (limited preview) Google82±1limited preview2026-09
23Gemini 3.8 Flash Google77±1api $1.50/M2026-09
32Gemini 3.7 Flash Google76±1api $1.50/M2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #224 of 393 * · 0.0% of its score · 1 test

Beats Granite 4.2 8B, Mistral Small 4 (119B-A6.5B), Olmo 3.1 32B Think here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
MixInstruct GPT-Rank (LLM-Blender, mean rank among 11 instruction-tuned LLMs, ChatGPT pairwise judge)8.8100 mean rank (1-11) ✎█···▄··▄···▄ beats 0/4avg 6.444 mean rank (1-11)▇ 0.0%blunt · noisy
math██·█···█···█ #1 of 5 * · 13% of its score · 1 test
testresultvs the others in its rundifficultycountswhy
MGSM lab4.9% ✎▃·▁·▃··▂▄█▄▄ beats 4/26hard · avg 22%▇▇▇▇▇ 13%too hard for it
other tests4 not in a task · 87% of its score
testresultvs the others in its rundifficultycountswhy
BIG-Bench Hard +143.9% ✎███··█·█·█·█ beats 6/6easy · avg 80%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 47%sharp · at its level · several runs
MMLU +154.5% ✎▄▄·▄···▄▄·██ beats 8/8easy · avg 82%▇▇▇▇▇▇▇▇▇ 23%at its level · several runs
Factual knowledge recall, Premium2k exact match (zero-shot)20.6% ✎▅····▃▅·██·█ beats 5/13medium · avg 59%▇▇▇▇ 10%typical
Factual knowledge recall, Premium2k exact match (10-shot)20.9% ✎▃▅▃··▅▃·█▃▃▅ beats 1/13easy · avg 67%▇▇▇ 7.4%at its level

$ where to run it

Other ways to run it

  1. Run it as published · needs ~45.07 GB · from huggingface.co/google (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text2text-generation', model='google/flan-t5-xxl', torch_dtype='auto', device_map='auto')('Answer the question: what is the capital of France?', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/google ↗

$ how the score was worked out · Scaling Instruction-Finetuned… 54%, InstructEval (DeCLaRe Lab) RE… 28%, Towards a Holistic Evaluation… 17% and 1 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
Scaling Instruction-Finetuned Language Models — authors' own runs · lab-reported3▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 54%CC BY 4.0
InstructEval (DeCLaRe Lab) README results table2▇▇▇▇▇▇▇▇ 28%MIT / Apache-2.0
Towards a Holistic Evaluation of LLMs on Factual Knowledge Recall (arXiv:2404.16164v1)2▇▇▇▇▇ 17%CC-BY-4.0
LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion (arXiv:2306.02561v3)1▇ 0%CC-BY-4.0
prior · pull toward the middle▇ 1%
weighted average= 33.3

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json