EleutherAIGPTNeoXForCausalLM

Pythia 160M — Hardware Requirements & GPU Compatibility

Chat

Pythia 160M is part of EleutherAI's Pythia training suite, a collection of models trained on the same data in the same order at multiple scales to enable rigorous scientific research into how language models learn. At 160 million parameters, it is the smallest model in the suite and runs on virtually any hardware. This model is primarily valuable for researchers studying scaling laws, training dynamics, and emergent capabilities across model sizes. EleutherAI released full training checkpoints, data, and code, making Pythia 160M one of the most transparent and reproducible models available for academic study.

213M params · dense2K contextApache 2.0Feb 20233.4M downloads66 likes4.8K quant downloads

Your hardware

Verdicts below are for the reference RTX 4090. Set yours to personalise the page.

Detection reads your browser's GPU name with no data sent anywhere.

S

Runs great on a 24 GB GPU

Runs from 8 GB (Q4_K_M · 2K)Best on 24 GB (Q4_K_M · ~4680 tok/s · 2K)Macs from 8 GB (Q4_K_M · ~512 tok/s)

Apache 2.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run Pythia 160M

One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.

Recommended

Ollama

one command · needs Ollama installed

$ollama run hf.co/mradermacher/pythia-160m-i1-GGUF:Q4_K_M

Q4_K_M fits fully on the RTX 4090 · 23.9 GB free · 2K context

Download: pythia-160m.i1-Q4_K_M.gguf0.11 GBmradermacher

install, OS, other quants

Ways to runOther ways

each opens step-by-step

Run in cloud

Runs on RTX 3080 Ti (12 GB) (11 GB headroom) · Q4_K_M

Generation speed: ~4236 tok/s; generation speed
Cost per 1M output tokens: $0.0080; per 1M output tokens

Deployon Vast.aiBest valuefrom $0.12/hr

Compare GPUs →

or

How Much VRAM Does Pythia 160M Need?

Select a quantization level; the listed hardware adheres to it.

Quantization	VRAM	+32K	+128K	File	Quality	On RTX 4090 (reference)
	0.1 GB	—	—	0.1 GB	2-bit · i-quant extreme compression — Importance-weighted 2-bit, extreme compression — significant quality loss	Fits: 2K context · ~10920 tok/s
	0.1 GB	—	—	0.1 GB	2-bit · i-quant extra small — Importance-weighted 2-bit, extra small	Fits: 2K context · ~9360 tok/s
	0.1 GB	—	—	0.1 GB	2-bit · i-quant small — Importance-weighted 2-bit, small	Fits: 2K context · ~9360 tok/s
	0.1 GB	—	—	0.1 GB	2-bit · i-quant medium — Importance-weighted 2-bit, medium	Fits: 2K context · ~8190 tok/s
	0.1 GB	—	—	0.1 GB	3-bit · i-quant — Importance-weighted 3-bit	Fits: 2K context · ~7280 tok/s
	0.1 GB	—	—	0.1 GB	3-bit · i-quant extra small — Importance-weighted 3-bit, extra small	Fits: 2K context · ~6552 tok/s
	0.1 GB	—	—	0.1 GB	3-bit small — 3-bit small quantization	Fits: 2K context · ~6552 tok/s
	0.1 GB	—	—	0.1 GB	2-bit · K-quant — 2-bit quantization with K-quant improvements	Fits: 2K context · ~6552 tok/s
	0.1 GB	—	—	0.1 GB	3-bit · i-quant small — Importance-weighted 3-bit, small	Fits: 2K context · ~6552 tok/s
	0.1 GB	—	—	0.1 GB	3-bit · i-quant medium — Importance-weighted 3-bit, medium	Fits: 2K context · ~5956 tok/s
	0.1 GB	—	—	0.1 GB	3-bit medium — 3-bit medium quantization	Fits: 2K context · ~5956 tok/s
	0.1 GB	—	—	0.1 GB	4-bit · legacy — 4-bit legacy quantization	Fits: 2K context · ~5460 tok/s
	0.1 GB	—	—	0.1 GB	3-bit large — 3-bit large quantization	Fits: 2K context · ~5460 tok/s
	0.1 GB	—	—	0.1 GB	4-bit · i-quant compact — Importance-weighted 4-bit, compact	Fits: 2K context · ~5040 tok/s
	0.1 GB	—	—	0.1 GB	4-bit small — 4-bit small quantization	Fits: 2K context · ~5040 tok/s
	0.1 GB	—	—	0.1 GB	4-bit medium — 4-bit medium quantization — most popular sweet spot	Fits: 2K context · ~4680 tok/s
	0.2 GB	—	—	0.1 GB	5-bit small — 5-bit small quantization	Fits: 2K context · ~4095 tok/s
	0.2 GB	—	—	0.1 GB	5-bit medium — 5-bit medium quantization — good quality/size tradeoff	Fits: 2K context · ~3854 tok/s
	0.2 GB	—	—	0.1 GB	6-bit, very good — 6-bit quantization, very good quality	Fits: 2K context · ~3448 tok/s
	0.2 GB	—	—	0.2 GB	8-bit, near-lossless — 8-bit quantization, near-lossless	Fits: 2K context · ~2849 tok/s
	0.5 GB	—	—	0.3 GB	16-bit · full — Full half-precision — baseline for inference	Fits: 2K context · ~1394 tok/s

Selected Q4_K_M

4-bit medium quantization

0.1GBat 2K context

Weights: 0.1 GB
KV cache: 0.0 GB
Overhead: 0.3 GB

Fits RTX 4090 (reference) · 23.9 GB free · ~4680 tok/s

≥50 tok/s needs ≥11 GB/s · RTX 5090 ~8320, RTX 3090 Ti ~4680

Which Hardware Can Run Pythia 160M?

Q4_K_M · 0.1 GB

Pythia 160M (Q4_K_M) requires 0.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Pythia 160M, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).

By VRAM tier · typical GPUs and Macs at Q4_K_M

Typical hardwareQuantVRAMtok/sContext that fits

8 GB · RTX 4060Q4_K_M · 0.1 GB · ~1263 tok/s · 2K0.1~12632K8 GB tier →

12 GB · RTX 3060 12GBQ4_K_M · 0.1 GB · ~1671 tok/s · 2K0.1~16712K12 GB tier →

16 GB · RTX 5060 Ti 16GBQ4_K_M · 0.1 GB · ~2080 tok/s · 2K0.1~20802K16 GB tier →

24 GB · RTX 4090referenceQ4_K_M · 0.1 GB · ~4680 tok/s · 2K0.1~46802K24 GB tier →

Mac 32 GB · Mac Mini M4Q4_K_M · 0.1 GB · ~600 tok/s · 2K0.1~6002KMacs →

Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 0.1 GB · ~2730 tok/s · 2K0.1~27302KMacs →

Alternatives to Pythia 160M

Similar models

1tok/s on RTX 4090

Model	Score	VRAM	tok/s	Why it is similar
MiniCPM V 4.7 1B DSparkOpenBMB	—	0.5 GB	~1489	same size

Specifications

Publisher: EleutherAI
Parameters: 213M
Architecture: GPTNeoXForCausalLM
Context Length: 2,048 tokens
Vocabulary Size: 50,304
Release Date: 2023-02-08
License: Apache 2.0

Resources

Hugging Face model card (opens in a new tab)

Frequently Asked Questions

How much VRAM does Pythia 160M need?

Pythia 160M requires 0.1 GB of VRAM at Q4_K_M, or 0.5 GB at FP16.

VRAM = Weights + KV Cache + Overhead

Weights = 213M × 4.8 bits ÷ 8 = 0.1 GB

Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.

VRAM usage by quantization

Q4_K_M

0.1 GB

Learn more about VRAM estimation →

What's the best quantization for Pythia 160M?

For Pythia 160M, Q4_K_M (0.1 GB) offers the best balance of quality and VRAM usage. Q5_K_S (0.2 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 0.1 GB.

VRAM requirement by quantization

IQ2_XXS

0.1 GB~53%

IQ3_XS

0.1 GB~73%

Q3_K_M

0.1 GB~83%

Q4_K_M ★

0.1 GB~89%

Q5_K_S

0.2 GB~92%

FP16

0.5 GB~100%

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Pythia 160M on a Mac?

Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Pythia 160M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.1 GB of usable unified memory (RAM minus macOS overhead).

Can I run Pythia 160M locally?

Yes — Pythia 160M can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Pythia 160M?

At Q4_K_M, Pythia 160M can reach ~34286 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~4680 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = (bandwidth GB/s ÷ model GB) × efficiency

Example: NVIDIA B200 → 8000 ÷ 0.14 × 0.65 = ~37143 tok/s

Estimated speed at Q4_K_M (0.1 GB)

NVIDIA B200

~37143 tok/s

NVIDIA GeForce RTX 4090

~4680 tok/s

NVIDIA B300

~37143 tok/s

AMD Instinct MI350X

~34286 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of Pythia 160M?

At Q4_K_M, the download is about 0.13 GB. The full-precision FP16 version is 0.43 GB. The smallest option (IQ2_XXS) is 0.06 GB.

Which GPUs can run Pythia 160M?

52 consumer GPUs can run Pythia 160M at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.

Which devices can run Pythia 160M?

163 devices with unified memory can run Pythia 160M at Q4_K_M (0.1 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.

Run Pythia 160M

0.1 GB VRAM at 2K context · fits the RTX 4090 (23.9 GB free)

Works on macOS, Windows and Linux.

Install Ollama (Linux)
$curl -fsSL https://ollama.com/install.sh | sh
Install Ollama (macOS / Windows)
Download and run the app; it installs the ollama command.
Download Ollama ↗
Run Q4_K_M
$ollama run hf.co/mradermacher/pythia-160m-i1-GGUF:Q4_K_M
Raise the context window
Type it inside the chat. For the API server instead, start it with OLLAMA_CONTEXT_LENGTH=2048 ollama serve.
$/set parameter num_ctx 2048
A larger context needs more memory than the weights alone.
Use the API
Ollama serves its API on http://localhost:11434.

Works on macOS, Windows and Linux.

Install LM Studio
Desktop app for macOS (Apple silicon), Windows and Linux.
lmstudio.ai ↗
Download the model
In the app press Ctrl/Cmd + Shift + M and paste https://huggingface.co/mradermacher/pythia-160m-i1-GGUF, then pick Q4_K_M. Or use the CLI (run LM Studio once first):
$lms get https://huggingface.co/mradermacher/pythia-160m-i1-GGUF@Q4_K_M
Open on Hugging Face ↗
Load and chat
Open the model loader, select the model and set Context Length to 2048. Leave GPU offload at maximum if it fits.
Start the local server (optional)
$lms server start

Works on macOS, Windows and Linux.

Install llama.cpp (macOS)
MacPorts, Nix and conda-forge also work.
$brew install llama.cpp
Install options ↗
Install llama.cpp (Windows)
$winget install llama.cpp
Install llama.cpp (Linux)
Download the prebuilt llama-*-bin-ubuntu-* archive for your GPU backend (CPU, Vulkan) from the releases page, unpack it and run the binaries from that folder. Homebrew (brew install llama.cpp) and conda-forge also work on Linux.
llama.cpp releases ↗
Start the server
Downloads the model on first run. The chat UI and an OpenAI-compatible API are at http://localhost:8080.
$llama-server -hf mradermacher/pythia-160m-i1-GGUF:Q4_K_M -c 2048 -ngl 99 --port 8080

Works on macOS, Windows and Linux.

Option A: Docker Model Runner
Docker Desktop: Settings > AI > Enable Docker Model Runner. Docker Engine on Linux: install the docker-model-plugin package.
$docker model run hf.co/mradermacher/pythia-160m-i1-GGUF:Q4_K_M
Model Runner docs ↗
The quant tag after the repo is how Hugging Face cards write it; if your Docker version rejects it, drop the tag and pick the quant by repo.
Option B: llama.cpp server image (NVIDIA)
Needs the NVIDIA Container Toolkit. Models are cached in the llama-cache volume. Open http://localhost:8080.
$docker run --gpus all -p 8080:8080 -v llama-cache:/root/.cache ghcr.io/ggml-org/llama.cpp:server-cuda -hf mradermacher/pythia-160m-i1-GGUF:Q4_K_M -c 2048 -ngl 99 --host 0.0.0.0 --port 8080
Image tags (cuda13, vulkan, rocm) ↗
Cache path inside the image is not in the docs; remove the -v flag if the volume complains.

Linux only.

Install vLLM
Linux only; on Windows run it inside WSL2. NVIDIA GPU shown; AMD ROCm, Intel and CPU builds have their own install pages.
$uv venv --python 3.12 --seed
$source .venv/bin/activate
$uv pip install vllm --torch-backend=auto
vLLM installation ↗
Serve the model
OpenAI-compatible API at http://localhost:8000/v1.
$vllm serve EleutherAI/pythia-160m --max-model-len 2048
Runs the original weights, not the Q4_K_M GGUF; needs ~0 GB of GPU memory.

Works on macOS, Windows and Linux.

Install Jan
jan.ai ↗
Find the model
Open the Hub tab, paste this model ID in Search, and choose Q4_K_M if offered.
$mradermacher/pythia-160m-i1-GGUF
Use the local API (optional)
Settings > Local API Server > Start Server. It listens on http://127.0.0.1:1337.

Rent GPUs hourly to run Q4_K_M right there. Pricing comes from partner marketplaces and varies.

Available GPU rentals for Q4_K_M, prioritized by value
GPU	Speed	$/1M tokens	Rent
RTX 3080 Ti (12 GB)CheapestCheapest11 GB free	~4236 tok/s	$0.0080	Deploy on Vast.aifrom $0.12/hr (opens in new tab)	12 GB · 11 GB free · ~4236 tok/s · $0.0080 / 1M tokens
RTX 3060 12GB11 GB free	~1671 tok/s	$0.010	Deploy on Vast.aifrom $0.061/hr (opens in new tab)	11 GB free · ~1671 tok/s · $0.010 / 1M tokens
RTX 4060 (8 GB)7 GB free	~1263 tok/s	$0.019	Deploy on Vast.aifrom $0.086/hr (opens in new tab)	8 GB · 7 GB free · ~1263 tok/s · $0.019 / 1M tokens
RTX A4000 (16 GB)15 GB free	~2080 tok/s	$0.012	Deploy on Vast.aifrom $0.088/hr (opens in new tab)	16 GB · 15 GB free · ~2080 tok/s · $0.012 / 1M tokens
RTX 4070 SUPER (12 GB)11 GB free	~2340 tok/s	$0.012	Deploy on Vast.aifrom $0.10/hr (opens in new tab)	12 GB · 11 GB free · ~2340 tok/s · $0.012 / 1M tokens
RTX PRO 4500 (32 GB)31 GB free	~4160 tok/s	$0.023	Deploy on RunPodfrom $0.34/hr (opens in new tab)	32 GB · 31 GB free · ~4160 tok/s · $0.023 / 1M tokens
RTX 4090 (24 GB)23 GB free	~4680 tok/s	$0.020	Deploy on RunPodfrom $0.34/hr (opens in new tab)	24 GB · 23 GB free · ~4680 tok/s · $0.020 / 1M tokens
A40 (48 GB)47 GB free	~3231 tok/s	$0.030	Deploy on RunPodfrom $0.35/hr (opens in new tab)	48 GB · 47 GB free · ~3231 tok/s · $0.030 / 1M tokens
L4 (24 GB)23 GB free	~1393 tok/s	$0.088	Deploy on RunPodfrom $0.44/hr (opens in new tab)	24 GB · 23 GB free · ~1393 tok/s · $0.088 / 1M tokens

Compare GPUs →