whatmodel.sh / Eurus-2-7B-PRIME

Eurus-2-7B-PRIME

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

Eurus-2-7B-PRIME is an open-weights language model from PRIME-RL, released December 2024. Its published size is 7.62B parameters and a 4k-token context window. Licence: apache-2.0.

Catalogue identity notes

Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2024-12-31). PRIME RL checkpoint (Qwen2.5-Math-7B base) from Tsinghua/UIUC et al. 7,615,616,512 parameters stored in FP32 (30.46 GB safetensors).

44.1±2.3#456 of 719 LLMsfrom 4 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of Eurus-2-7B-PRIME, out of 828 with a score, then more from PRIME-RL.

#modelscorewhereRAMreleased
543Qwen2-Audio 7B Instruct Alibaba44±3local~6.5GB2024-07
544Gemma 3 4B Google44±1api $0.063/M +local~3.8GB2025-03
545Eurus-2-7B-PRIME PRIME-RL44±2local~6.4GB2024-12
546Llama 3.1 Tulu 3 8B Ai244±1local~6.2GB2024-11
547Qwen2.5-VL 3B Instruct Alibaba44±1local~3.5GB2025-01
more from PRIME-RL
696Eurus-2-7B-SFT PRIME-RL35±3local~6.4GB2024-12

see all LLMs · find the ones that fit your machine

$ what it's good at

math█··········█ #1 of 2 · 65% of its score · 3 tests
testresultvs the others in its rundifficultycountswhy
AIME 2024 lab26.7% ✎▁▁▁▂▁▃▁▅███▃ beats 14/85medium · avg 42%▇▇▇▇▇▇▇▇▇▇▇ 29%sharp · at its level · small run
MATH-500 lab79.2% ✎▁▃▂▃▅▇▅▇█▅▂▁ beats 21/84very easy · avg 87%▇▇▇▇▇▇▇ 19%small run
Minerva MATH lab38.6% ✎▄·▂▂·██▆·▆▄▄ beats 10/21medium · avg 59%▇▇▇▇▇▇▇ 18%at its level · small run
other tests1 not in a task · 32% of its score
testresultvs the others in its rundifficultycountswhy
GPQA Diamond33.9% ✎▁▁▁▁▃▄▆▄▅▇█▅ beats 25/208hard · avg 40%▇▇▇▇▇▇▇▇▇▇▇▇▇ 32%sharp

$ 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 Eurus-2-7B-PRIME · needs ~4.7 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/Eurus-2-7B-PRIME-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/Eurus-2-7B-PRIME-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 Eurus-2-7B-PRIME · needs ~4.7 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/Eurus-2-7B-PRIME-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 Eurus-2-7B-PRIME · needs ~4.7 GB · from huggingface.co/bartowski
    chat in the terminal
    ollama run hf.co/bartowski/Eurus-2-7B-PRIME-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 Eurus-2-7B-PRIME · needs ~8.1 GB, Mac only · from huggingface.co/WaveCut
    chat in the terminal
    mlx_lm.chat --model WaveCut/PRIME-RL_Eurus-2-7B-PRIME-Q8-mlx
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model WaveCut/PRIME-RL_Eurus-2-7B-PRIME-Q8-mlx

Other ways to run it

  1. Run it as published · needs ~37.08 GB, NVIDIA only · from huggingface.co/PRIME-RL (official)
    vllm serve PRIME-RL/Eurus-2-7B-PRIME

Official weights and documentation: huggingface.co/PRIME-RL ↗

$ how the score was worked out · PRIME-RL Eurus-2-7B-PRIME mod… 65%, Epoch AI — internal evaluatio… 32%

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
PRIME-RL Eurus-2-7B-PRIME model card, Evaluation table (own models) · lab-reported3▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 65%Factual vendor-reported score (not licensed content)
Epoch AI — internal evaluations1▇▇▇▇▇▇▇▇▇▇ 32%CC BY 4.0
prior · pull toward the middle▇ 2%
weighted average= 44.1

$ 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