whatmodel.sh / OpenLLaMA 3B v2

OpenLLaMA 3B v2

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

OpenLLaMA 3B v2 is an open-weights language model from OpenLM Research, released July 2023. Licence: apache-2.0.

Catalogue identity notes

Maker repo openlm-research/open_llama_3b_v2 @bce5d60: licence tag apache-2.0. Released 2023-07: month the maker's Hugging Face repo was created (createdAt 2023-07-16). Pretrained base checkpoint (not an instruct/chat version).

24.5±1.0#676 of 719 LLMsfrom 12 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of OpenLLaMA 3B v2, out of 828 with a score, then more from OpenLM Research.

#modelscorewhereRAMreleased
779ChatGLM2-6B Zhipu AI (THUDM)25±2local~15GB2023-06
780GPT-J 6B EleutherAI25±1local~14GB2021-06
781OpenLLaMA 3B v2 OpenLM Research25±1local~4.2GB2023-07
782OPT-6.7B Meta24±1local~16GB2022-05
783RWKV-4-Raven 14B RWKV (BlinkDL)24±2comparison only2023-04
more from OpenLM Research
748OpenLLaMA 7B v2 OpenLM Research29±1local~9.1GB2023-07

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 Challenge36.1% ✎▁·▁··▁▁▇▃█▆▃ beats 10/26easy · avg 85%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 38%sharp
WinoGrande62.9% ✎▁····▁··▁▃█▄ beats 7/26easy · avg 89%▇▇▇▇▇ 12%at its level
BoolQ65.7% ✎▂▃▂▅▆▂▅▆█▂▂▂ beats 9/26easy · avg 91%▇▇▇▇ 10%at its level
ARC Easy63.5% ✎▁▁·▁▁▄▅▄█▅▁▃ beats 10/26easy · avg 88%▇▇▇ 8.3%at its level
SciQ92.4% ✎▁·▁·▁·▃▅▃█▄▁ beats 12/26very easy · avg 98%▇▇▇ 7.8%typical
HellaSwag70.0% ✎▃···▁▇▅█▇▃▁▁ beats 12/26very easy · avg 92%▇▇▇ 7.0%at its level
PIQA76.7% ✎▁·▁··▁▁▃█▅█▂ beats 14/26easy · avg 88%▇▇ 5.2%blunt · at its level
LAMBADA (OpenAI version), last-word accuracy66.7% ✎▁··▂▁▁▁▂▆█▅▃ beats 8/26very easy · avg 89%▇ 3.4%blunt · at its level
4 more tests
testresultvs the others in its rundifficultycountswhy
OpenBookQA26.0% ✎█·▁▃▃·▁▃▅▇▃▄ beats 7/26medium · avg 53%▇ 2.8%too hard for it
Social IQA41.2% ✎▁···▁▃█▁▁▁·▁ beats 7/26medium · avg 59%▇ 2.5%blunt
Open PT-LLM Leaderboard average (9 Portuguese tasks)30.6% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 33/276medium · avg 73%▇ 1.1%noisy
TruthfulQA MC234.6% ✎█·█·█▄···▄·█ beats 3/9medium · avg 61%▇ 0.6%blunt · at its level

$ where to run it

llama.cpp · recommended

  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 OpenLLaMA 3B v2 · needs ~2.6 GB · from huggingface.co/lstgst
    chat in the terminal
    llama cli -hf lstgst/open_llama_3b_v2-Q4_K_M-GGUF:q4_k_m
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf lstgst/open_llama_3b_v2-Q4_K_M-GGUF:q4_k_m

LM Studio

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

    download it (macOS / Windows / Linux) ↗

  2. Run OpenLLaMA 3B v2 · needs ~2.6 GB · from huggingface.co/lstgst

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/lstgst/open_llama_3b_v2-Q4_K_M-GGUF@q4_k_m -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run OpenLLaMA 3B v2 · needs ~2.6 GB · from huggingface.co/lstgst
    chat in the terminal
    ollama run hf.co/lstgst/open_llama_3b_v2-Q4_K_M-GGUF:q4_k_m

Other ways to run it

  1. Run it as published · needs ~6.85 GB · from huggingface.co/openlm-research (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='openlm-research/open_llama_3b_v2', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/openlm-research ↗

$ how the score was worked out · Stability AI — original Stabl… 98%, 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 outputs11▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 98%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 1%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 24.5

$ 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