whatmodel.sh / INTELLECT-3

INTELLECT-3

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

INTELLECT-3 is an open-weights language model from Prime Intellect, released November 2025. Its published size is 106B parameters (12B active per token). Licence: mit.

Catalogue identity notes

Comparison-only. Arena entry 'intellect-3' (text). Maker repo PrimeIntellect/INTELLECT-3 @ff39d4a: licence tag mit. Parameters: card: 'a 106B (A12B) parameter Mixture-of-Experts reasoning model post-trained from GLM-4.5-Air-Base'. Released 2025-11: Prime Intellect blog 'INTELLECT-3' dated Nov 26th, 2025.

64.8±1.1#145 of 719 LLMsfrom 9 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of INTELLECT-3, out of 828 with a score.

#modelscorewhereRAMreleased
165Qwen3.5 Flash Alibaba65±1api $0.17/M2026-02
166ChatGPT-4o (2025-03-26) OpenAI65±2comparison only2025-03
167INTELLECT-3 Prime Intellect65±1local~48GB2025-11
168Nemotron 3 Super 120B-A12B NVIDIA65±1api $0.28/M +local~73GB2026-03
169GLM-4.5 Z.ai65±1api $1.00/M +local~58GB2025-07

see all LLMs · find the ones that fit your machine

$ what it's good at

coding▁▁▁▃▆██▇▆▄▃▁ #153 of 458 * · 4.6% of its score · 1 test

Beats Gemini 2.5 Pro, Gemini 3.1 Flash-Lite, Qwen3-Coder-Next (80B-A3B) here, though they rank higher overall. Trails Ministral 3 8B Reasoning 2512, Phi-4-reasoning 14B, DeepSeek V3 (Dec '24) here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LiveCodeBench v6 (Aug 2024–May 2025) lab69.3% ✎▄▄·▄·····█·▄ beats 3/5medium · avg 45%▇▇ 4.6%at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #153 of 393 * · 12% of its score · 1 test

Trails LFM2.5 1.2B Thinking, LFM2.5 8B-A1B, LFM2.5 2.6B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LMArena (text)1355.8 ✎▁▁▁▂▃▃▃▆▆▇█▃ beats 200/379avg 1268▇▇▇▇▇ 12%blunt
math▁▂▃▅▆▅▄█▄▁▁▁ #52 of 398 · 24% of its score · 3 tests

Beats LongCat 2.0 (1.6T-A48B), Qwen3.6 Max Preview, Claude Sonnet 4.6 here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
AIME 2024 lab90.8% ✎█···███·█··█ beats 5/5medium · avg 49%▇▇▇▇▇ 13%sharp
AIME 2025 lab88.0% ✎█·█···███··█ beats 5/5medium · avg 43%▇▇▇▇ 11%sharp
MATH-500 lab98.1% ✎▄······▄▄▄·█ beats 5/5very easy · avg 86%▇ 1.3%maxed out · noisy
other tests4 not in a task · 59% of its score
testresultvs the others in its rundifficultycountswhy
GPQA Diamond lab74.4% ✎█··██·██···█ beats 2/5hard · avg 42%▇▇▇▇▇▇▇▇▇ 23%sharp · at its level
Medmarks-V mean win rate (19 verifiable medical benchmarks, 71 configurations)0.5660 ✎▁▄▃▃▇▄█▃▅▁▁▁ beats 24/30avg 0.423▇▇▇▇▇▇▇ 18%sharp
Humanity's Last Exam lab14.6% ✎▄····▄·█·▄·▄ beats 2/5very hard · avg 4.3%▇▇▇▇▇ 12%typical
MMLU-Pro lab81.9% ✎▁▁·▁▁▁▃▂▄▅█▁ beats 139/188medium · avg 51%▇▇▇ 6.4%typical

$ 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 INTELLECT-3 · needs ~71.36 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/PrimeIntellect_INTELLECT-3-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/PrimeIntellect_INTELLECT-3-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 INTELLECT-3 · needs ~71.36 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/PrimeIntellect_INTELLECT-3-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 INTELLECT-3 · needs ~44.71 GB · from huggingface.co/bartowski
    chat in the terminal
    ollama run hf.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF:Q2_K

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 INTELLECT-3 · needs ~60.1 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/INTELLECT-3-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/INTELLECT-3-4bit

Official weights and documentation: huggingface.co/PrimeIntellect ↗

$ how the score was worked out · Prime Intellect INTELLECT-3 m… 70%, Medmarks: A Comprehensive Ope… 18%, LMArena leaderboard dataset 12%

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 Intellect INTELLECT-3 model card: Evaluation table (Environments Hub runs) · lab-reported7▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 70%MIT
Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks — authors' own runs of 61 models1▇▇▇▇▇ 18%CC BY 4.0
LMArena leaderboard dataset1▇▇▇ 12%CC BY 4.0
prior · pull toward the middle▇ 1%
weighted average= 64.8

$ 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