[whatmodel.sh](../../README.md) / INTELLECT-3
# INTELLECT-3
Prime Intellect · released 2025-11 · 106B params (12B active) · mit
**score 64.2** ±1.1 · #154 of 732 · from 9 benchmark results across 3 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks INTELLECT-3 among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁▁▁▃▅▇█●▅▄▃▁ #160 of 469* 3.9% 1
chat & writing ▁▁▁▃▆▄▆█●▇▇▃ #160 of 402* 12% 1
math ▁▁▃▅█▆▄█●▂▁▁ #59 of 431 22% 3
coding · #160 of 469 · 3.9% of its score
Beats [Qwen3-Coder-Next (80B-A3B)](qwen3-coder-next.md), [Gemini 2.5 Pro](gemini-2-5-pro.md), [Gemini 3.1 Flash-Lite](gemini-3-1-flash-lite.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 8B Reasoning 2512](ministral-3-8b-reasoning.md), [Phi-4-reasoning 14B](phi-4-reasoning.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LiveCodeBench v6 (Aug 2024–May 20… 69.3% ▄▄·▄·····●·▄ beats 3/5 medium · avg 45% 3.9% at its level
chat & writing · #160 of 402 · 12% of its score
Trails [LFM2.5 1.2B Thinking](lfm2-5-1-2b-thinking.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md), [LFM2.5 2.6B](lfm2-5-2-6b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LMArena (text) 1355.5 ▁▁▁▂▃▃▃▆●▇█▃ beats 200/385 avg 1260 12% blunt
math · #59 of 431 · 22% of its score
Beats [Qwen3.6 Max Preview](qwen3-6-max-preview.md), [Claude Sonnet 4.6](claude-sonnet-4-6.md), [DeepSeek V4 Pro (Preview, Apr 2026)](deepseek-v4-pro-preview.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
AIME 2024 (maker) 90.8% █···███·█··● beats 5/5 medium · avg 50% 11% sharp
AIME 2025 (maker) 88.0% █·█···███··● beats 5/5 medium · avg 42% 10% sharp
MATH-500 (maker) 98.1% ▄······▄▄▄·● beats 5/5 very easy · avg 86% 1.1% maxed out · noisy
other tests (4)
test result vs the others in its run difficulty counts why
GPQA Diamond (maker) 74.4% █··█●·██···█ beats 2/5 hard · avg 41% 23% sharp · at its level
Medmarks-V mean win rate (19 veri… 0.5660 ▁▄▃▃▇▄█▃●▁▁▁ beats 24/30 avg 0.418 22% sharp
Humanity's Last Exam (maker) 14.6% ▄····▄·●·▄·▄ beats 2/5 very hard · avg 4.1% 11% typical
MMLU-Pro (maker) 81.9% ▁▁·▁▁▁▃▂▄▅●▁ beats 139/188 medium · avg 50% 6.2% typical
source results share of the score
Prime Intellect INTELLECT-3 model card:… 7 66% ▇▇▇▇▇▇▇▇
Medmarks: A Comprehensive Open-Source L… 1 22% ▇▇▇
LMArena leaderboard dataset 1 12% ▇
prior: pull toward the middle 0% ▇
weighted average = 64.2
Licences: Prime Intellect INTELLECT-3 model card: Evaluation table ([MIT](https://huggingface.co/PrimeIntellect/INTELLECT-3/blob/ff39d4a4688989f3f28868923d030c28e1b7d81c/README.md)) · Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks ([CC BY 4.0](../../licenses/paper-2605.01417v1-CC-BY-4.0.txt)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run INTELLECT-3 (Q4_K_M, needs ~71.36 GB, from [huggingface.co/bartowski](https://huggingface.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF/tree/1827cd1b39b3f3ced9319c94079d71bad75ee8c3))
```sh
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)
```sh
llama serve -hf bartowski/PrimeIntellect_INTELLECT-3-GGUF:Q4_K_M
```
Other versions: Q2_K (~44.71 GB) `llama cli -hf bartowski/PrimeIntellect_INTELLECT-3-GGUF:Q2_K`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run INTELLECT-3 (Q4_K_M, needs ~71.36 GB, from [huggingface.co/bartowski](https://huggingface.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF/tree/1827cd1b39b3f3ced9319c94079d71bad75ee8c3))
```sh
lms get https://huggingface.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q2_K (~44.71 GB) `lms get https://huggingface.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF@Q2_K -y` · 4bit (~60.1 GB, Mac only) `lms get https://huggingface.co/mlx-community/INTELLECT-3-4bit -y` · 8bit (~113.6 GB, Mac only) `lms get https://huggingface.co/mlx-community/INTELLECT-3-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run INTELLECT-3 (Q2_K, needs ~44.71 GB, from [huggingface.co/bartowski](https://huggingface.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF/tree/1827cd1b39b3f3ced9319c94079d71bad75ee8c3))
```sh
ollama run hf.co/bartowski/PrimeIntellect_INTELLECT-3-GGUF:Q2_K
```
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run INTELLECT-3 (4bit, needs ~60.1 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/INTELLECT-3-4bit/tree/8b6e6c7751c39f87f871451f243d1b0d0f5fbade))
```sh
mlx_lm.chat --model mlx-community/INTELLECT-3-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/INTELLECT-3-4bit
```
Other versions: 8bit (~113.6 GB, Mac only) `mlx_lm.chat --model mlx-community/INTELLECT-3-8bit`
Model repository and documentation: [huggingface.co/PrimeIntellect](https://huggingface.co/PrimeIntellect/INTELLECT-3)
[interactive version](https://whatmodel.sh/model/intellect-3) · [JSON](https://whatmodel.sh/api?format=json&all=1)