whatmodel.sh / Qwen2 57B-A14B Instruct

Qwen2 57B-A14B Instruct

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

Qwen2 57B-A14B Instruct is an open-weights language model from Alibaba, released June 2024. Its published size is 57.41B parameters (14B active per token) and a 33k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo Qwen/Qwen2-57B-A14B-Instruct @50896d6: licence tag apache-2.0. 57,408,658,944 parameters from the Hugging Face safetensors metadata; active 14B per token (Qwen2 card: '#Activated Params ... 14B' for 57B-A14B). Released 2024-06: month the maker's Hugging Face repo was created (createdAt 2024-06-04).

47.6±1.7#387 of 719 LLMsfrom 5 benchmark results across 4 sources

$ how it compares

The LLMs ranked either side of Qwen2 57B-A14B Instruct, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
462Granite 4.0 H Tiny (7B-A1B) IBM48±1local~5.6GB2025-09
463DeepSeek-Coder 7B Instruct v1.5 DeepSeek48±1local~5.9GB2024-01
464Qwen2 57B-A14B Instruct Alibaba48±2local~35GB2024-06
465Qwen2.5 Coder 3B Instruct Alibaba48±1local~3.3GB2024-11
466DeepSeek-Coder 6.7B Instruct DeepSeek48±1local~5.8GB2023-10
more from Alibaba
18Qwen3.8 Max (2.4T-A95B) Alibaba78±1api $3.00/M +local~380GB2026-08
34Qwen3.8 Max (0902) Alibaba76±1api $3.00/M2026-09
37Qwen3.8-Flash-Next Alibaba75±1local~59GB2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests5 not in a task · 99% of its score
testresultvs the others in its rundifficultycountswhy
BigCodeBench (Instruct) lab35.7% ✎▃▂·▆▆██▆▅▃▃▃ beats 17/33medium · avg 41%▇▇▇▇▇▇▇▇▇▇▇▇▇ 33%at its level
BigCodeBench (Complete) lab46.5% ✎▂▁▁▄▂▅▇█▆▂▂▂ beats 44/58medium · avg 50%▇▇▇▇▇▇▇▇▇▇▇▇▇ 33%at its level
Polish EQ-Bench v2 (SpeakLeash)57.6% ✎▁··▁▁▂▂▆▅▆█▆ beats 36/70medium · avg 64%▇▇▇▇▇▇ 15%at its level
Open Persian LLM Leaderboard v2 average (21 tasks)56.5% ✎▂▄▂▃█▄▃▆▃▅▂▂ beats 46/81medium · avg 58%▇▇▇▇▇ 12%blunt · at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)75.5% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 229/276medium · avg 73%▇▇ 5.9%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 Qwen2 57B-A14B Instruct · needs ~34.9 GB · from huggingface.co/Qwen (official)
    chat in the terminal
    llama cli -hf Qwen/Qwen2-57B-A14B-Instruct-GGUF:q4_k_m
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf Qwen/Qwen2-57B-A14B-Instruct-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 Qwen2 57B-A14B Instruct · needs ~34.9 GB · from huggingface.co/Qwen (official)

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/Qwen/Qwen2-57B-A14B-Instruct-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 Qwen2 57B-A14B Instruct · needs ~34.9 GB · from huggingface.co/Qwen (official)
    chat in the terminal
    ollama run hf.co/Qwen/Qwen2-57B-A14B-Instruct-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 Qwen2 57B-A14B Instruct · needs ~32.3 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Qwen2-57B-A14B-Instruct-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Qwen2-57B-A14B-Instruct-4bit

Other ways to run it

  1. Run it as published · needs ~114.82 GB · from huggingface.co/Qwen (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='Qwen/Qwen2-57B-A14B-Instruct', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · BigCodeBench paper (Zhuo et a… 66%, SpeakLeash Polish EQ-Bench le… 15%, Open Persian LLM Leaderboard… 12% 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
BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v4 · lab-reported2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 66%CC BY 4.0
SpeakLeash Polish EQ-Bench leaderboard1▇▇▇▇▇ 15%Apache-2.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)1▇▇▇ 12%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇▇ 6%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 47.6

$ 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