whatmodel.sh / GPT-NeoX-20B

GPT-NeoX-20B

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

GPT-NeoX-20B is an open-weights language model from EleutherAI, released February 2022. Its published size is 20.55B parameters and a 2k-token context window. Licence: apache-2.0.

28.0±0.8#657 of 719 LLMsfrom 20 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of GPT-NeoX-20B, out of 828 with a score, then more from EleutherAI.

#modelscorewhereRAMreleased
754OLMo 7B Allen Institute for AI28±1local~30GB2024-02
755Falcon 40B Instruct TII28±2local~27GB2023-05
756GPT-NeoX-20B EleutherAI28±1local~14GB2022-02
757LLaMA 13B Meta28±1comparison only2023-02
758Koala 13B UC Berkeley (BAIR)28±1comparison only2023-04
more from EleutherAI
777Pythia 12B EleutherAI25±1local~9.5GB2023-02
780GPT-J 6B EleutherAI25±1local~14GB2021-06
787Pythia 6.9B EleutherAI23±1local~6.1GB2023-02

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests12 not in a task · 100% of its score
testresultvs the others in its rundifficultycountswhy
ARC Challenge +140.8% ✎▁·▁··▁▁▇▂█▆▃ beats 15/26easy · avg 85%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 38%sharp · several runs
WinoGrande +166.1% ✎▁····▁··▁▂█▄ beats 15/26easy · avg 89%▇▇▇▇▇ 13%at its level · several runs
HellaSwag +171.4% ✎▃···▁▇▅█▇▃▁▁ beats 14/26very easy · avg 92%▇▇▇ 7.9%at its level · several runs
ARC Easy +168.7% ✎▁▁·▁▁▄▅▄█▅▁▃ beats 15/26easy · avg 88%▇▇▇ 7.3%at its level · several runs
BoolQ69.5% ✎▂▃▂▅▆▂▅▆█▂▂▂ beats 12/26easy · avg 91%▇▇▇ 6.7%at its level
SciQ +193.1% ✎▁·▁·▁·▃▅▃█▄▁ beats 14/26very easy · avg 98%▇▇ 5.8%several runs
OpenBookQA +129.8% ✎█·▁▁▃·▁▃▅▇▃▄ beats 13/26medium · avg 52%▇▇ 5.7%too hard for it · several runs
TriviaQA lab25.9% ✎█···█·███··█ beats 4/5easy · avg 72%▇▇ 4.4%at its level
4 more tests
testresultvs the others in its rundifficultycountswhy
PIQA +177.4% ✎▁·▁··▁▁▃█▅█▂ beats 16/26easy · avg 88%▇▇ 4.4%blunt · at its level · several runs
LAMBADA (OpenAI version), last-word accuracy +172.0% ✎▁··▂▁▁▁▂▆█▅▃ beats 20/26very easy · avg 89%▇ 3.6%blunt · at its level · several runs
Social IQA42.0% ✎▁···▁▃█▁▁▁·▁ beats 11/26medium · avg 59%▇ 1.7%blunt
Open PT-LLM Leaderboard average (9 Portuguese tasks)28.3% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 19/276medium · avg 73%▇ 0.8%noisy

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run GPT-NeoX-20B · needs ~13.1 GB · from huggingface.co/mradermacher
    chat in the terminal
    llama cli -hf mradermacher/gpt-neox-20b-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf mradermacher/gpt-neox-20b-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run GPT-NeoX-20B · needs ~13.1 GB · from huggingface.co/mradermacher

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/mradermacher/gpt-neox-20b-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run GPT-NeoX-20B · needs ~13.1 GB · from huggingface.co/mradermacher
    chat in the terminal
    ollama run hf.co/mradermacher/gpt-neox-20b-GGUF:Q4_K_M

MLX (Mac)

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run GPT-NeoX-20B · needs ~11.6 GB, Mac only · from huggingface.co/nhe-ai
    chat in the terminal
    mlx_lm.chat --model nhe-ai/gpt-neox-20b-mlx-4Bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model nhe-ai/gpt-neox-20b-mlx-4Bit

Other ways to run it

  1. Run it as published · needs ~41.29 GB · from huggingface.co/EleutherAI (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='EleutherAI/gpt-neox-20b', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/EleutherAI ↗

$ how the score was worked out · Stability AI — original Stabl… 67%, GPT-NeoX-20B: An Open-Source… 32%, Open Portuguese LLM Leaderboa… 1%

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
Stability AI — original StableLM zero-shot evaluation outputs10▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 67%Apache-2.0
GPT-NeoX-20B: An Open-Source Autoregressive Language Model — authors' own runs · lab-reported9▇▇▇▇▇▇▇▇▇ 32%CC BY 4.0
Open Portuguese LLM Leaderboard (archived)1▇ 1%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 28.0

$ 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