Skip to content

GPT Neox 20B on RTX 5080: VRAM, speed and fit

GPT Neox 20B at 4-bit runs on the RTX 5080 with 2K of context at about 51 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on RTX 5080 · 4-bit (Q4_K_M) · Ollama / LM Studio · Chat

GPU memory
14.5 GB
of 16.0 GB
Context
2K
auto
Speed
51 tok/s
Very fast
First reply
0.8 s
2K prompt

Where the memory goes

Model
11.7 GB
Context
2.1 GB
Overhead
751 MB
OS
0.0 GB
Free
1.5 GB
Memory
14.5 GB of 16.0 GB
Context
2K tokens · auto
Speed
51 tok/s · very fast
First reply
0.8 s for a 2K prompt
Running cost
$6.48/mo at home · $29/mo on a cloud RTX 2000 Ada

Validation: Validated

What GPT Neox 20B on RTX 5080 was checked against
CheckTargetResultStatus
Fits in GPU memory16.0 GB usable14.5 GB neededPass
Headroom5% free1.5 GB free (9%)Pass
Context for chat2K or more2KPass
Writing speed15 tok/s or more51 tok/sPass
First replyunder 3 s, 2K prompt0.8 sPass

Targets: a reply you can read along with, a first reply the use case tolerates, and 5% of memory left free. Calculated from published specs, not benchmarked.

The setup

Model
GPT Neox 20B
Parameters
20.7B
Precision
4-bit (Q4_K_M)
Context
2K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
RTX 5080
GPU can use
16.0 GB of 16.0 GB

Running cost

At home
$6.48/mo
296 W at the wall
Renting instead
$29/mo
RTX 2000 Ada at $0.24/h

4 h a day, power at $0.18 per kWh, cloud at RunPod on-demand prices as of Nov 21, 2026.

How to run it

llama.cpp
llama-server -hf mradermacher/gpt-neox-20b-i1-GGUF
vLLM
vllm serve EleutherAI/gpt-neox-20b

GPT Neox 20B on similar machines

  • Radeon RX 9070 XTRuns
  • RTX 4070 Ti SuperRuns
  • RTX 4080 SuperRuns
  • RTX 5070 TiRuns
  • Radeon RX 7900 XTXRuns
  • RTX 3090Runs
  • RTX 4090Runs
  • Radeon AI PRO R9700Runs

All machines: GPT Neox 20B VRAM requirements. Everything that runs here: what can the RTX 5080 run?

Other models that run on RTX 5080

Questions

Can GPT Neox 20B run on RTX 5080?

GPT Neox 20B at 4-bit runs on the RTX 5080 with 2K of context at about 51 tokens a second.

Report VR-19LF18G, calculated Nov 22, 2026 from published specs, not benchmarked.