SAVRN
Search Contact SAVRN

Open-weight model · Text generation

pythia-160m

by EleutherAI EleutherAI/pythia-160m

The Pythia Scaling Suite is a collection of models developed to facilitate interpretability research (see paper). It contains two sets of eight models of sizes 70M, 160M, 410M, 1B, 1.4B, 2.8B, 6.9B, and 12B.

Parameters213M
Context2,048
Weights750.0 MB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads3.5M

Runs On

What it takes to serve pythia-160m (213M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 0.4 GB 0.5 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 0.2 GB 0.3 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 0.1 GB 0.1 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

SAVRN's Notes on pythia-160m

Nobody stands up an accelerator for a model that fits in half a gigabyte, which is the point of Pythia-160m. EleutherAI built the Pythia Scaling Suite for interpretability research: eight sizes from 70M to 12B, trained on the Pile in the same order, with 154 intermediate checkpoints each. It needs 0.5 GB at 16-bit; 8-bit takes that to 0.3 GB and 4-bit to 0.1 GB. The price table bottoms out at one MI300X with 192 GB at $1.85 an hour on-demand, a floor rather than a fit. We would run it in spare memory.

Nothing on the license side slows you down: Apache 2.0 permits commercial use, modification and redistribution if you keep the license and copyright notices and state significant changes. Two checks before building on it: the 2,048 token context, and the parameter count, 213M on the listing against the 160m in the name.

SAVRN Research, 2026-09-18

Model Card

By EleutherAI, published under apache-2.0, revision 50f5173d932e.

The Pythia Scaling Suite is a collection of models developed to facilitate interpretability research (see paper). It contains two sets of eight models of sizes 70M, 160M, 410M, 1B, 1.4B, 2.8B, 6.9B, and 12B. For each size, there are two models: one trained on the Pile, and one trained on the Pile after the dataset has been globally deduplicated. All 8 model sizes are trained on the exact same data, in the exact same order. We also provide 154 intermediate checkpoints per model, hosted on Hugging Face as branches.

The Pythia model suite was deliberately designed to promote scientific research on large language models, especially interpretability research. Despite not centering downstream performance as a design goal, we find the models match or exceed the performance of similar and same-sized models, such as those in the OPT and GPT-Neo suites.

Read the full model card (1,371 words)

Configuration

Architecture
GPTNeoXForCausalLM
Context length (tokens)
2,048
Layers
12
Hidden size
768
Feed-forward size
3,072
Attention heads
12
Vocabulary size
50,304
Stored precision
float16
Model type
gpt_neox

Identity and Version

Repository
EleutherAI/pythia-160m
Publisher
EleutherAI
Task
Text generation
Modality
Text
Library
transformers
Parameters
213M parameters
Languages
en
Revision
50f5173d932e8e61f858120bcb800b97af589f46
First published
2023-02-08
Last updated
2023-07-09

Files and Weights

8 files, 752.2 MB in total. The weights are 2 files totalling 750.0 MB in bin, safetensors.

Weights2 files · 750.0 MB
Configuration2 files · 668 B
Tokenizer2 files · 2.1 MB
Documentation1 file · 13.6 KB
Repository1 file · 1.5 KB
Every file
FileTypeSizeSHA-256
model.safetensorsWeights375.0 MB 29d2e457a664
pytorch_model.binWeights375.0 MB 8d856725c4a8
config.jsonConfiguration569 B —
special_tokens_map.jsonConfiguration99 B —
README.mdDocumentation13.6 KB —
.gitattributesRepository1.5 KB —
tokenizer.jsonTokenizer2.1 MB —
tokenizer_config.jsonTokenizer396 B —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
750.0 MB
Download from EleutherAI

Released by EleutherAI through its official repository on Hugging Face. Read the license.

Built From

Memory Requirements

PrecisionWeights in memory
As published750.0 MB
16-bit0.4 GB
8-bit0.2 GB
4-bit0.1 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Compare pythia-160m

Questions About pythia-160m

How much GPU memory does pythia-160m need?

About 0.5 GB at 16-bit and 0.1 GB at 4-bit: the weights (213M parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run pythia-160m on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use pythia-160m commercially?

Yes. pythia-160m is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

What is pythia-160m's context length?

2,048 tokens, from the maximum position embeddings in its published configuration.

Similar Models

Model · Text generation

CasualSwarms

Convergent Intelligence

SAGI is a novel causal language model that integrates swarm intelligence dynamics with transformer architecture. The model treats cognition as a dynamic, adaptive system where multiple internal "agents" collaborate through differentiable routing, trust mechanisms, and shared memory. The enhancements were integrated with the existing AGI system through: 1. Compatibility Layer: Ensuring new components work with existing AGI Core 2. Unified State Representation: Combining enhanced capabilities with existing state 3. Enhanced Continuous Learning: Upgrading the learning system with new capabilities 4. Performance Monitoring: Tracking improvements through validation systems - Successfully…

Open weights apache-2.0 170M parameters 1,024 tokens transformers
View model

Model · Text generation

Ru-Small-Instruct

LongTime

Ru-Small-Instruct — экспериментальная компактная русскоязычная языковая модель класса SLM (Small Language Model) с объемом параметров ~0.2B (~165M). Разработана с упором на суверенность весов (Zero-Fingerprint): модель обучена с нуля без заимствования базовых чекпоинтов у сторонних корпоративных сетей (Llama 3 от Meta, Qwen от Alibaba, Mistral). Модель предназначена для исследований локального инференса, работы на маломощном оборудовании, CPU и мобильных чипах, где критичны нулевая задержка (Time-To-First-Token) и полная независимость весов. Для компактной модели в 165M параметров, обученной на одном домашнем GPU за 48 часов, способность держать роль, грамотно формулировать сложные термины…

Open weights mit 165M parameters 512 tokens transformers
View model

This model is a fine-tuned derivative of google/gemma-3-270m, adapted using the Convergent Intelligence sparse fine-tuning setup originally tested on Liquid Foundation Models. The checkpoint was trained on reasoning-style English examples from angrygiraffe/claude-opus-4.6-4.7-reasoning-8.7k using a targeted adaptation strategy and the custom CIxOpt optimizer framework. The goal of this model is to test whether a compact Gemma 3 270M backbone can be shaped toward reasoning-style text generation through selective parameter participation rather than broad full-model modification. This is an experimental research checkpoint intended for evaluation, local testing, optimizer research, and…

Open weights gemma 268M parameters 262,144 tokens transformers
View model

This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Open weights apache-2.0 268M parameters 32,768 tokens transformers
View model

Model · Text generation

vlt5-base-keywords

VoiceLab.ai

Results on demo model (different generation method, one model per language): Keywords generated with vlT5-base-keywords: encoder-decoder architecture, vlT5, keyword generation, scientific articles corpus The biggest advantage is the transferability of the vlT5 model, as it works well on all domains and types of text. The downside is that the text length and the number of keywords are similar to the training data: the text piece of an abstract length generates approximately 3 to 5 keywords. It works both extractive and abstractively. Longer pieces of text must be split into smaller chunks, and then propagated to the model. The model was trained on a POSMAC corpus. Polish Open Science…

Open weights cc-by-4.0 275M parameters transformers
View model

Model · Text generation

Haidass-Translate-143M

DALab

English | 中文 A 143M-parameter bidirectional Chinese↔English translation model, instruction-tuned on the Haidass1.5-143M base — the strongest zh⇄en translator at this scale among general chat-architecture models. Drafter-143M: a control model with identical configuration, data and training recipe, except that it starts from random initialization instead of the pretrained base — used to quantify the contribution of base-model pretraining. OPUS-MT models are single-directional — one independent 78M model per direction; "-" marks directions a model does not serve. The same models re-evaluated on FLORES+ devtest (released 2026; zero overlap with dev): devtest sentences do not overlap with dev.…

Open weights apache-2.0 143M parameters 4,096 tokens
View model