whatmodel.sh / BLOOMZ 7.1B
BLOOMZ 7.1B
↗ share
+ contribute
BLOOMZ 7.1B is an open-weights language model from BigScience, released November 2022. Its published size is 7.07B parameters. Licence: bigscience-bloom-rail-1.0.
Catalogue identity notes
BLOOM 7.1B multitask-finetuned on xP3 (arXiv 2211.01786); checkpoint bigscience/bloomz-7b1. Not the -mt / -p3 variants.
Model information sources
Score within the LMArena (text) comparison set only.
$ how it compares
The LLMs ranked either side of BLOOMZ 7.1B, out of 877 with a score, then more from BigScience.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 794 |  EuroLLM 1.7B Instruct | 29±3 | local | ||
| 795 |  Granite 4.0 H 350M | 29±1 | local | ||
| 796 | BLOOMZ 7.1B | 29±2 | comparison only | ||
| 797 | BLOOM 176B | 29±1 | local | ||
| 798 |  Gemma 2B IT | 29±1 | local | ||
| more from BigScience | |||||
| 743 | T0 11B | 35±3 | comparison only | ||
| 759 | BLOOMZ 176B | 33±2 | comparison only | ||
| 808 | BLOOMZ 3B | 27±2 | comparison only | ||
$ what it's good at
coding█···▄·▄·▄··▄ #2 of 6 * · 70% of its score · 1 test
other tests6 not in a task · 28% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| DiBiMT lexical-ambiguity translation accuracy (5-direction average) | 43.6% ✎ | ██···██···██ beats 1/5 | 10% |
| WinoGrande lab | 55.4% ✎ | █··▁▃·▁··▃·▃ beats 10/12 | 8.3% |
| Conic10K mathQA (zero-shot CoT) | 0.0% ✎ | █··▄·······▄ beats 1/3 | 5.4% |
| COPA lab | 83.0% ✎ | ▃··█▅▃▃▅·▃▃▃ beats 10/12 | 3.3% |
| IndoMMLU (zero-shot) | 38.0% ✎ | ▅▃▅··▃···▃▃█ beats 10/10 | 0.8% |
| Story Cloze Test lab | 89.1% ✎ | █·▂··▃··▂▂▃▃ beats 10/12 | 0.7% |
$ where to run it
No verified API or download recorded yet · add one
Official weights and documentation: huggingface.co/bigscience ↗
$ how the score was worked out · Crosslingual Generalization t… 82%, Towards Effective Disambiguat… 10%, Conic10K: A Challenging Math… 5% 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
| source | results | share of the score |
|---|---|---|
 Crosslingual Generalization through Multitask Finetuning — authors' own runs · lab-reported | 4 | 82% |
| Towards Effective Disambiguation for Machine Translation with Large Language Models — authors' own runs | 1 | 10% |
| Conic10K: A Challenging Math Problem Understanding and Reasoning Dataset — authors' own runs | 1 | 5% |
| Large Language Models Only Pass Primary School Exams in Indonesia: A Comprehensive Test on IndoMMLU — authors' own runs | 1 | 1% |
| prior · pull toward the middle | 2% | |
| weighted average | = 28.9 |
$ improve this page
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