whatmodel.sh / Llama 3 8B (base)

Llama 3 8B (base)

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

Llama 3 8B (base) is an open-weights language model from Meta, released April 2024. Its published size is 8.03B parameters. Licence: llama3.

Catalogue identity notes

Maker repo meta-llama/Meta-Llama-3-8B @8cde5ca: licence tag llama3. 8,030,261,248 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2024-04: month the maker's Hugging Face repo was created (createdAt 2024-04-17). Pretrained base checkpoint (not an instruct/chat version). Gated repository (licence acceptance required on Hugging Face).

35.6±0.7#596 of 719 LLMsfrom 9 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of Llama 3 8B (base), out of 828 with a score, then more from Meta.

#modelscorewhereRAMreleased
688Tulu 2 7B Allen Institute for AI36±2local~5.8GB2023-11
689xDAN-L1-Chat-RL-v1 xDAN-AI36±1local~6.1GB2023-12
690Llama 3 8B (base) Meta36±1local~6.6GB2024-04
691Mistral 7B v0.1 (base) Mistral AI36±0local~5.8GB2023-09
692Qwen2.5 1.5B Instruct Alibaba36±1local~2.4GB2024-09
more from Meta
24Muse Spark 1.3 Meta77±1api $2.00/M2026-09
40Muse Spark 1.2 Meta75±1api $2.00/M2026-08
43Muse Spark 1.1 Meta75±1api $2.00/M2026-07

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #417 of 458 · 83% of its score · 3 tests

Beats DeepSeek R1 Distill Qwen 7B, Qwen3.5 2B, DeepSeek R1 Distill Llama 8B here, though they rank higher overall. Trails Llama 3.2 1B Instruct, SmolLM2 1.7B Instruct, Granite 3B Code Base (2K) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
HumanEval33.5% ✎▁·▁▁▃▄▆▅█▃▁▁ beats 18/99easy · avg 79%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 46%sharp · at its level
HumanEval+29.3% ✎▁·▁▁▁▃▄▄▄█▃▁ beats 18/99easy · avg 74%▇▇▇▇▇▇▇▇▇▇▇▇ 31%sharp · at its level
MBPP+51.6% ✎▁···▁▁▂▅▇█▆▂ beats 21/75easy · avg 68%▇▇ 6.2%at its level
other tests6 not in a task · 16% of its score
testresultvs the others in its rundifficultycountswhy
MedQA (USMLE, 4 options)59.7% ✎▁▁··▄▁▁··▄▃█ beats 16/17easy · avg 76%▇▇ 5.3%sharp · at its level
BigCodeBench (Complete)28.8% ✎▂▁▁▄▂▅▇█▆▂▂▂ beats 9/58medium · avg 50%▇▇ 5.0%at its level
MedMCQA57.5% ✎▂▂··▄▄▂▂·▄██ beats 16/17medium · avg 68%▇ 2.9%at its level
Open Persian LLM Leaderboard v2 average (21 tasks)47.6% ✎▂▄▂▃█▄▃▆▃▅▂▂ beats 23/81medium · avg 58%▇ 2.1%blunt · at its level
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy82.3% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 77/146easy · avg 83%▇ 0.8%at its level · noisy
PubMedQA74.8% ✎▂·▂▆·▄▂▄▄▂█▂ beats 14/17medium · avg 77%▇ 0.4%blunt · at its level

$ 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 Llama 3 8B (base) · needs ~4.9 GB · from huggingface.co/QuantFactory
    chat in the terminal
    llama cli -hf QuantFactory/Meta-Llama-3-8B-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf QuantFactory/Meta-Llama-3-8B-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 Llama 3 8B (base) · needs ~4.9 GB · from huggingface.co/QuantFactory

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/QuantFactory/Meta-Llama-3-8B-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 Llama 3 8B (base) · needs ~4.9 GB · from huggingface.co/QuantFactory
    chat in the terminal
    ollama run hf.co/QuantFactory/Meta-Llama-3-8B-GGUF:Q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Llama 3 8B (base) · needs ~5.3 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Meta-Llama-3-8B-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Meta-Llama-3-8B-4bit

Other ways to run it

  1. Run it as published · needs ~16.06 GB · from huggingface.co/meta-llama
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='meta-llama/Meta-Llama-3-8B', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Weights and documentation: huggingface.co/meta-llama ↗

$ how the score was worked out · EvalPlus / EvalPerf published… 83%, Open Medical-LLM Leaderboard… 9%, BigCodeBench paper (Zhuo et a… 5% and 2 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
EvalPlus / EvalPerf published evaluation runs3▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 83%Apache-2.0
Open Medical-LLM Leaderboard results (HF dataset openlifescienceai/results)3▇▇▇ 9%Apache-2.0
BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v41▇ 5%CC BY 4.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)1▇ 2%Apache-2.0
EuroEval leaderboards (results collection), English generative leaderboard1▇ 1%MIT
prior · pull toward the middle▇ 0%
weighted average= 35.6

$ 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