What are the hardware requirements for INTELLECT 3 5bit?

INTELLECT 3 5bit requires approximately 73.4 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed INTELLECT 3 5bit and how large is it?

INTELLECT 3 5bit is developed by mlx-community, a model with 106.9b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does INTELLECT 3 5bit perform on standard benchmarks?

INTELLECT 3 5bit has the following published scores: LMSYS ELO 1356. Compare against reference models on this page or on the LLM Explorer leaderboards.

INTELLECT 3 5bit by mlx-community — VRAM 73.4GB, 128K context

Name: INTELLECT 3 5bit
Author: mlx-community

INTELLECT 3 5bit is an open-source language model by mlx-community. Features: 106.9b LLM, VRAM: 73.4GB, Context: 128K, License: mit, Quantized, LLM Explorer Score: 0.35, ELO: 1356.

5-bit 5bit Agentic Base model:primeintellect/inte... Base model:quantized:primeinte... Conversational Custom code En Glm4 moe Mixture-of-experts Mlx Prime-intellect Prime-rl Quantized Reasoning Region:us Safetensors Sharded Tensorflow Verifiers

Model Card on HF 🤗: https://huggingface.co/mlx-community/INTELLECT-3-5bit

INTELLECT 3 5bit Benchmarks

LMSys ELO 1356 vs 1272 (so35) 6.6%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

INTELLECT 3 5bit Parameters and Internals

LLM Name	INTELLECT 3 5bit
Repository 🤗	https://huggingface.co/mlx-community/INTELLECT-3-5bit
Base Model(s)	INTELLECT 3 PrimeIntellect/INTELLECT-3
Model Size	106.9b
Required VRAM	73.4 GB
Updated	2026-08-11
Maintainer	mlx-community
Model Type	glm4_moe
Model Files	5.4 GB: 1-of-14 5.0 GB: 2-of-14 5.3 GB: 3-of-14 5.3 GB: 4-of-14 5.3 GB: 5-of-14 5.4 GB: 6-of-14 5.4 GB: 7-of-14 5.4 GB: 8-of-14 4.9 GB: 9-of-14 5.3 GB: 10-of-14 5.3 GB: 11-of-14 5.3 GB: 12-of-14 5.4 GB: 13-of-14 4.7 GB: 14-of-14
Supported Languages	en
Quantization Type	5bit
Model Architecture	Glm4MoeForCausalLM
License	mit
Context Length	131072
Model Max Length	131072
Transformers Version	4.56.1
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|endoftext\|>
Vocabulary Size	151552

Rank the INTELLECT 3 5bit capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to INTELLECT 3 5bit

Best Alternatives	Context / RAM	Downloads	Likes
INTELLECT 3 4bit	128K / 60.4 GB	14	3
GLM 4.5 Air 2bit	128K / 33.4 GB	107	4
GLM 4.5 Air 8bit	128K / 114.1 GB	2726	11
GLM 4.5 Air 4bit	128K / 60.4 GB	1261	28
GLM 4.5 Air 3bit	128K / 46.6 GB	621	32
GLM 4.5 Air 6bit	128K / 86.7 GB	201	3
GLM 4.5 Air 3bit DWQ V2	128K / 46.6 GB	82	4
GLM 4.5 Air 3bit DWQ	128K / 53.3 GB	50	4
GLM 4.5 Air 2bit DWQ	128K / 40.2 GB	28	2
INTELLECT 3	128K / 194.3 GB	6275	216

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/INTELLECT-3-5bit.