whatmodel.sh / Aya 23 8B

Aya 23 8B

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

Aya 23 8B is an open-weights language model from Cohere For AI, released May 2024. Its published size is 8.03B parameters and a 8k-token context window. Licence: cc-by-nc-4.0.

Catalogue identity notes

Maker repo CohereLabs/aya-23-8B @89da1a0: licence tag cc-by-nc-4.0. 8,028,033,024 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2024-05: month the maker's Hugging Face repo was created (createdAt 2024-05-19). The repo was previously published as CohereForAI/aya-23-8B (the Hugging Face API redirects that id here); leaderboard rows may use the old id. Ollama tags aya:8b-23-q4_K_M, aya:8b-23-q8_0 are the library's instruct/chat builds of this checkpoint (tag names on the pinned tags page); download sizes are estimated from the parameter count.

42.6±2.8#492 of 719 LLMsfrom 4 benchmark results across 4 sources

$ how it compares

The LLMs ranked either side of Aya 23 8B, out of 828 with a score, then more from Cohere For AI.

#modelscorewhereRAMreleased
580Qwen2.5 3B Instruct Alibaba43±1local~3.3GB2024-09
581DeepSeek-Coder 1.3B Instruct DeepSeek43±1local~2.4GB2023-10
582Aya 23 8B Cohere For AI43±3local~6.2GB2024-05
583Phi-3-small-8k-instruct Microsoft43±2local~17GB2024-05
584CodeGemma 7B IT (v1.0) Google43±1local~7.1GB2024-03
more from Cohere For AI
429Aya 23 35B Cohere For AI49±3local~22GB2024-05

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests4 not in a task · 97% of its score
testresultvs the others in its rundifficultycountswhy
Polish EQ-Bench v2 (SpeakLeash)45.4% ✎▁··▁▁▂▂▆▅▆█▆ beats 16/70medium · avg 64%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 41%at its level
Open Persian LLM Leaderboard v2 average (21 tasks)48.5% ✎▂▄▂▃█▄▃▆▃▅▂▂ beats 24/81medium · avg 58%▇▇▇▇▇▇▇▇▇▇▇▇ 30%blunt · at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)66.3% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 123/276medium · avg 73%▇▇▇▇▇▇ 15%at its level · noisy
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy71.3% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 39/146easy · avg 83%▇▇▇▇ 10%at its level · 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 Aya 23 8B · needs ~5.1 GB · from huggingface.co/lmstudio-community
    chat in the terminal
    llama cli -hf lmstudio-community/aya-23-8B-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf lmstudio-community/aya-23-8B-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 Aya 23 8B · needs ~5.1 GB · from huggingface.co/lmstudio-community

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/lmstudio-community/aya-23-8B-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 Aya 23 8B · needs ~5.32 GB · from ollama.com
    chat in the terminal
    ollama run aya:8b-23-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 Aya 23 8B · needs ~4.5 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/aya-23-8B-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/aya-23-8B-4bit

Official weights and documentation: huggingface.co/CohereLabs ↗

$ how the score was worked out · SpeakLeash Polish EQ-Bench le… 41%, Open Persian LLM Leaderboard… 30%, Open Portuguese LLM Leaderboa… 15% 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

sourceresultsshare of the scorelicence
SpeakLeash Polish EQ-Bench leaderboard1▇▇▇▇▇▇▇▇▇▇▇▇ 41%Apache-2.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)1▇▇▇▇▇▇▇▇▇ 30%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇▇▇▇▇ 15%Apache-2.0
EuroEval leaderboards (results collection), English generative leaderboard1▇▇▇ 10%MIT
prior · pull toward the middle▇ 3%
weighted average= 42.6

$ 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