[whatmodel.sh](../../README.md) / BLOOM 7.1B
# BLOOM 7.1B
BigScience · released 2022-07 · 7.07B params · bigscience-bloom-rail-1.0
**score 20.6** ±0.7 · #700 of 732 · from 56 benchmark results across 13 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks BLOOM 7.1B among every model on that task or test, worst to best.
task vs the others rank counts tests
coding █··█·█·█●··█ #2 of 6* 5.1% 1
math ▁●▃▅█▆▄█▄▂▁▁ #418 of 431* 5.5% 1
coding · #2 of 6 · 5.1% of its score
test result vs the others in its run difficulty counts why
HumanEval (maker) 7.7% ▄···▄·▄·●··▄ beats 4/5 easy · avg 79% 5.1% sharp · too hard for it
math · #418 of 431 · 5.5% of its score
Beats [Flan-T5 Large](flan-t5-large.md), [OPT-66B](opt-66b.md), [Flan-T5 Base](flan-t5-base.md) here, though they rank higher overall. Trails [GPT-Neo 125M](gpt-neo-125m.md), [OPT-125M](opt-125m.md), [GPT-2 XL (1.5B)](gpt-2-xl.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K 1.9% ▄▄█●···▄·▄·▄ beats 4/7 easy · avg 65% 5.5% sharp · too hard for it · several runs
other tests (27)
test result vs the others in its run difficulty counts why
ARC Challenge 33.5% ▁·▁··▁▁●▂█▆▃ beats 5/26 easy · avg 82% 22% sharp · several runs
WinoGrande 64.6% ▁····▁··▁▂●▄ beats 11/26 easy · avg 88% 9.1% several runs
ARC Easy 57.4% ▁▁·▁▁●▅▄█▅▁▃ beats 4/26 easy · avg 83% 8.8% blunt · at its level · several runs
HellaSwag 62.3% ▃···▁●▅█▇▃▁▁ beats 5/26 easy · avg 81% 7.8% at its level · several runs
PIQA 72.7% ▁·▁··▁●▃█▅█▂ beats 3/26 easy · avg 87% 5.6% blunt · at its level · several runs
BoolQ 62.8% ▂▃▂●▆▂▅▆█▂▂▂ beats 4/26 easy · avg 88% 5.5% at its level · several runs
COPA (maker) 69.0% ▃··█▅▃▃●·▃▃▃ beats 8/12 very easy · avg 95% 4.1% at its level · several runs
TriviaQA 26.3% ▁··▁█▃▃●▄▅▄█ beats 12/28 medium · avg 45% 3.8% too hard for it · several runs
SciQ 90.1% ▁·▁·▁·▃●▃█▄▁ beats 7/26 very easy · avg 97% 2.9% blunt
OpenBookQA 25.2% ●·▁▁▃·▁▃▅▇▃▄ beats 2/26 easy · avg 77% 2.9% several runs
LAMBADA (OpenAI version), last-wo… 57.6% ▁··▂▁●▁▂▆█▅▃ beats 4/26 very easy · avg 87% 2.7% blunt · at its level · several runs
Natural Questions (closed book) 8.1% ▁·▁·▁▂▁●·▅▂█ beats 7/28 medium · avg 48% 2.5% several runs
Story Cloze Test 72.0% ▂▂▄▂·▂·●·▂█▂ beats 8/14 easy · avg 89% 2.4% blunt · at its level · several runs
CoQA 68.8% ▄····▄▄·▄●·█ beats 4/6 very easy · avg 92% 1.5% at its level
Winograd Schema Challenge (WSC273) 81.3% ▄▄·▄·▄··●··█ beats 5/7 very easy · avg 93% 1.3% blunt · at its level
DiBiMT lexical-ambiguity translat… 41.0% ●█···██···██ beats 0/5 easy · avg 74% 1.1% blunt · at its level
Factual knowledge recall, Premium… 3.2% ●····▃▅·██·█ beats 0/13 medium · avg 52% 1.0% too hard for it
Factual knowledge recall, Premium… 18.1% ●▅▃··▅▃·█▃▃▅ beats 0/13 medium · avg 61% 1.0% typical
RACE-high 36.6% ▄···█··▄●··█ beats 4/6 medium · avg 51% 0.8% blunt
Social IQA 42.1% ▁···▁▃●▁▁▁·▁ beats 14/26 hard · avg 56% 0.8% blunt
DROP 9.9% ▄▄···█·●·▄·▄ beats 4/6 hard · avg 34% 0.7% blunt
Open PT-LLM Leaderboard average (… 28.9% ●▁▁▁▁▁▁▁▂█▅▁ beats 27/276 medium · avg 72% 0.4% noisy
TruthfulQA MC2 38.9% ▅●█▆▁▂▅···▁▁ beats 6/31 medium · avg 57% 0.3% blunt · at its level · several runs
MenatQA (all factors), exact matc… 9.6% ▂····●··▂·█▄ beats 1/8 medium · avg 36% 0.2% blunt
HaluEval QA hallucination detecti… 58.1% █▁▁·▁▁·▁●▁▁▁ beats 23/27 hard · avg 66% 0.1% too hard for it · noisy
WebQuestions (closed book) 2.3% █···▄··▄▄▄·● beats 6/6 very hard · avg 3.5% 0.0% blunt · too hard for it
LogiQA (harness) 20.3% ●··········▁ beats 9/14 very hard · avg 25% 0.0% too hard for it · noisy
source results share of the score
Stability AI 10 30% ▇▇▇▇
EleutherAI lm-evaluation-harness result… 17 30% ▇▇▇▇
GPTQ: Accurate Post-Training Quantizati… 5 14% ▇▇
UPRISE: Universal Prompt Retrieval for … 9 12% ▇
BLOOM: A 176B-Parameter Open-Access Mul… 1 5% ▇
Hallucinations Leaderboard results 4 3% ▇
INFORM : Information eNtropy based mult… 1 2% ▇
Towards a Holistic Evaluation of LLMs o… 2 2% ▇
Towards Effective Disambiguation for Ma… 1 1% ▇
Crosslingual Generalization through Mul… 3 1% ▇
Open Portuguese LLM Leaderboard 1 0% ▇
MenatQA: A New Dataset for Testing the … 1 0% ▇
From Test-Taking to Test-Making: Examin… 1 0% ▇
prior: pull toward the middle 0% ▇
weighted average = 20.6
Licences: Stability AI ([Apache-2.0](../../licenses/stablelm-original-eval-json-Apache-2.0.txt)) · EleutherAI lm-evaluation-harness results folder ([MIT](../../licenses/lm-eval-harness-results-2023-05-MIT.txt)) · GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers ([CC BY 4.0](../../licenses/paper-2210.17323v2-gptq-CC-BY-4-0.txt)) · UPRISE: Universal Prompt Retrieval for Improving Zero-Shot Evaluation ([CC BY 4.0](../../licenses/acl-2023-emnlp-main-758-CC-BY-4-0.txt)) · BLOOM: A 176B-Parameter Open-Access Multilingual Language Model ([CC BY 4.0](../../licenses/paper-2211.05100v4-CC-BY-4.0.txt)) · Hallucinations Leaderboard results ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · INFORM : Information eNtropy based multi-step reasoning FOR large language Models ([CC BY 4.0](../../licenses/acl-2023-emnlp-main-216-CC-BY-4-0.txt)) · Towards a Holistic Evaluation of LLMs on Factual Knowledge Recall ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · Towards Effective Disambiguation for Machine Translation with Large Language Models ([CC BY 4.0](../../licenses/acl-2023-wmt-1-44-CC-BY-4-0.txt)) · Crosslingual Generalization through Multitask Finetuning ([CC BY 4.0](../../licenses/acl-2023-acl-long-891-CC-BY-4-0.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · MenatQA: A New Dataset for Testing the Temporal Comprehension and Reasoning Abilities of Large Language Models ([CC BY 4.0](../../licenses/acl-2023-findings-emnlp-100-CC-BY-4-0.txt)) · From Test-Taking to Test-Making: Examining LLM Authoring of Commonsense Assessment Items ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run BLOOM 7.1B (Q4_K_M, needs ~5.3 GB, from [huggingface.co/mradermacher](https://huggingface.co/mradermacher/bloom-7b1-GGUF/tree/375f264e00bb9e4565ea017a6c82ae5cb5ae819b))
```sh
llama cli -hf mradermacher/bloom-7b1-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf mradermacher/bloom-7b1-GGUF:Q4_K_M
```
Other versions: Q8_0 (~8.6 GB) `llama cli -hf mradermacher/bloom-7b1-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run BLOOM 7.1B (Q4_K_M, needs ~5.3 GB, from [huggingface.co/mradermacher](https://huggingface.co/mradermacher/bloom-7b1-GGUF/tree/375f264e00bb9e4565ea017a6c82ae5cb5ae819b))
```sh
lms get https://huggingface.co/mradermacher/bloom-7b1-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~8.6 GB) `lms get https://huggingface.co/mradermacher/bloom-7b1-GGUF@Q8_0 -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run BLOOM 7.1B (Q4_K_M, needs ~5.3 GB, from [huggingface.co/mradermacher](https://huggingface.co/mradermacher/bloom-7b1-GGUF/tree/375f264e00bb9e4565ea017a6c82ae5cb5ae819b))
```sh
ollama run hf.co/mradermacher/bloom-7b1-GGUF:Q4_K_M
```
Other versions: Q8_0 (~8.6 GB) `ollama run hf.co/mradermacher/bloom-7b1-GGUF:Q8_0`
### Other ways to run it
1. Run it as published (fp16, needs ~14.14 GB, from [huggingface.co/bigscience](https://huggingface.co/bigscience/bloom-7b1/tree/6232703e399354503377bf59dfbb8397fd569e4a))
```sh
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='bigscience/bloom-7b1', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
```
Model repository and documentation: [huggingface.co/bigscience](https://huggingface.co/bigscience/bloom-7b1)
[interactive version](https://whatmodel.sh/model/bloom-7b1) · [JSON](https://whatmodel.sh/api?format=json&all=1)