INTELLECT-1 is the first collaboratively trained 10 billion parameter language model trained from scratch on 1 trillion tokens of English text and code.

This is a base model. Please use the INTELLECT-1-Instruct for chat use case.

INTELLECT-1 was trained on up to 14 concurrent nodes distributed across 3 continents, with contributions from 30 independent community contributors providing compute. The training code utilizes the prime framework, a scalable distributed training framework designed for fault-tolerant, dynamically scaling, high-perfomance training on unreliable, globally distributed workers. The key abstraction that allows dynamic scaling is the ElasticDeviceMesh which manages dynamic global process groups for fault-tolerant communication across the internet and local process groups for communication within a node. The model was trained using the DiLoCo algorithms with 100 inner steps. The global all-reduce was done with custom int8 all-reduce kernels to reduce the communication payload required, greatly reducing the communication overhead by a factor 400x.

For more detailed technical insights, please refer to our technical paper.

Note: You must add a BOS token at the beginning of each sample. Performance may be impacted otherwise.

Usage

python

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

torch.set_default_device("cuda")
model = AutoModelForCausalLM.from_pretrained("PrimeIntellect/INTELLECT-1")
tokenizer = AutoTokenizer.from_pretrained("PrimeIntellect/INTELLECT-1")

input_text = "What is the Metamorphosis of Prime Intellect about?"
input_ids = tokenizer.encode(input_text, return_tensors="pt")
output_ids = model.generate(input_ids, max_length=50, num_return_sequences=1)
output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)

print(output_text)

Example text generation pipeline

python

import torch
from transformers import pipeline
torch.set_default_device("cuda")

pipe = pipeline("text-generation", model="PrimeIntellect/INTELLECT-1")
print(pipe("What is prime intellect ?"))

Model Details

Compute Contributors: Prime Intellect, Arcee AI, kotaro, skre_0, marlo, rodeo, Herb, Olas, superchillen, Hugging Face, mev_pete, 0xfr_, dj, primeprimeint1234, Marco Giglio, realtek, Hyperbolic, hecataeus, NWO, Virtual Machine, droll, SemiAnalysis, waiting_, toptickcrypto, sto, Johannes, washout_segment_0b, klee
Release Date: 29 Nov 2024
Model License: Apache 2.0

Technical Specifications

Parameter	Value
Parameter Size	10B
Number of Layers	42
Number of Attention Heads	32
Hidden Size	4096
Context Length	8192
Vocabulary Size	128256

Training Details:

Dataset: 55% fineweb-edu, 10% fineweb, 20% Stack V1, 10% dclm-baseline, 5% open-web-math
Tokens: 1 Trillion
Optimizer: Diloco/LocalSGD - Inner Optimizer: AdamW, Outer Optmizer: Nesterov SGD

Performance on benchmarks

Base Models:

Model	Size	Tokens	MMLU	GPQA	GSM8K	ARC-C	Hellaswag
INTELLECT	10B	1T	37.5	26.12	8.1	52.13	72.26
MPT-7B	7B	1T	26.8	25.67	8.3	46.67	77.41
Falcon-7B	7B	1.5T	26.2	23.66	4.9	47.61	78.23
Pythia-12B	12B	300B	26.5	24.33	4.09	40.61	68.83
LLM360-Amber	7B	1.3T	24.5	27.01	4.32	42.75	74.08
LLaMA-7B	7B	1T	35.1	23.21	9.7	50.43	78.19
LLaMA-13B	13B	1T	46.9	26.34	17.3	56.14	81.05
LLaMA2-7B	7B	2T	45.3	25.89	13.5	54.10	78.64
LLaMA2-13B	13B	2T	54.8	25.67	24.3	59.81	82.58

Instruction-Tuned Models:

Model	Size	Tokens	MMLU	GPQA	GSM8K	ARC-C	Hellaswag
INTELLECT-Instruct	10B	1T	49.89	28.32	38.58	54.52	71.42
MPT-7B-Chat	7B	1T	36.29	26.79	8.26	51.02	75.88
Falcon-7B-Instruct	7B	1.5T	25.21	26.34	4.93	45.82	70.61
LLM360-AmberChat	7B	1.4T	36.02	27.23	6.14	43.94	73.94
LLaMA2-7B-Chat	7B	2T	47.20	28.57	23.96	53.33	78.69
LLaMA2-13B-Chat	13B	2T	53.51	28.35	37.15	59.73	82.47

Citations

If you use this model in your research, please cite it as follows:

@article{jaghouar2024intellect,
  title={INTELLECT-1 Technical Report.},
  author={Jaghouar, Sami and Ong, Jack Min and Basra, Manveer and Obeid, Fares and Straube, Jannik and Keiblinger, Michael and Bakouch, Elie and Atkins, Lucas and Panahi, Maziyar and Goddard, Charles and Ryabinin, Max and Hagemann, Johannes},
  journal={arXiv preprint},
  year={2024}
}