Skip to content

LightOnOCR 2 1B on RTX 3080 10 GB: VRAM, speed and fit

LightOnOCR 2 1B at 8-bit runs on the RTX 3080 10 GB with 16K of context at about 351 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on RTX 3080 10 GB · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
3.5 GB
of 10.0 GB
Context
16K
auto
Speed
351 tok/s
157 when full
First reply
0.1 s
2K prompt

Where the memory goes

Model
1.0 GB
Context
1.8 GB
Overhead
790 MB
OS
0.0 GB
Free
6.5 GB
Memory
3.5 GB of 10.0 GB
Context
16K tokens · auto
Speed
351 tok/s · 157 when full
First reply
0.1 s for a 2K prompt
Running cost
$5.95/mo at home · $29/mo on a cloud RTX 2000 Ada

Validation: Validated

What LightOnOCR 2 1B on RTX 3080 10 GB was checked against
CheckTargetResultStatus
Fits in GPU memory10.0 GB usable3.5 GB neededPass
Headroom5% free6.5 GB free (65%)Pass
Context for chat16K or more16KPass
Writing speed15 tok/s or more351 tok/sPass
First replyunder 3 s, 2K prompt0.1 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
LightOnOCR 2 1B
Parameters
1B
Precision
8-bit (Q8_0)
Context
16K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
RTX 3080 10 GB
GPU can use
10.0 GB of 10.0 GB

Running cost

At home
$5.95/mo
272 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 22, 2026.

How to run it

llama.cpp
llama-server -hf noctrex/LightOnOCR-2-1B-GGUF
vLLM
vllm serve lightonai/LightOnOCR-2-1B

LightOnOCR 2 1B on similar machines

All machines: LightOnOCR 2 1B VRAM requirements. Everything that runs here: what can the RTX 3080 10 GB run?

Other models that run on RTX 3080 10 GB

Questions

Can LightOnOCR 2 1B run on RTX 3080 10 GB?

LightOnOCR 2 1B at 8-bit runs on the RTX 3080 10 GB with 16K of context at about 351 tokens a second.

Report VR-0I4V454, calculated Nov 22, 2026 from published specs, not benchmarked.