Skip to content

Medgemma 1.5 4B IT on RTX 3060 12 GB: VRAM, speed and fit

Medgemma 1.5 4B IT at 8-bit runs on the RTX 3060 12 GB with 128K of context at about 61 tokens a second.

Text generation · Chat · Ollama / LM Studio

Runs comfortably · on RTX 3060 12 GB · 8-bit (Q8_0) · Ollama / LM Studio · Chat

GPU memory
7.9 GB
of 12.0 GB
Context
128K
auto
Speed
61 tok/s
36 when full
First reply
0.5 s
2K prompt

Where the memory goes

Model
4.3 GB
Context
2.6 GB
Overhead
1.0 GB
OS
0.0 GB
Free
4.1 GB
Memory
7.9 GB of 12.0 GB
Context
128K tokens · auto
Speed
61 tok/s · 36 when full
First reply
0.5 s for a 2K prompt
Running cost
$3.98/mo at home · $29/mo on a cloud RTX 2000 Ada

Validation: Validated

What Medgemma 1.5 4B IT on RTX 3060 12 GB was checked against
CheckTargetResultStatus
Fits in GPU memory12.0 GB usable7.9 GB neededPass
Headroom5% free4.1 GB free (34%)Pass
Context for chat16K or more128KPass
Writing speed15 tok/s or more61 tok/sPass
First replyunder 3 s, 2K prompt0.5 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
Medgemma 1.5 4B IT
Parameters
4.3B
Precision
8-bit (Q8_0)
Context
128K, auto
Chats at once
1
Use
Chat
Runs in
Ollama / LM Studio
Hardware
RTX 3060 12 GB
GPU can use
12.0 GB of 12.0 GB

Running cost

At home
$3.98/mo
182 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 unsloth/medgemma-1.5-4b-it-GGUF
vLLM
vllm serve google/medgemma-1.5-4b-it

Medgemma 1.5 4B IT on similar machines

All machines: Medgemma 1.5 4B IT VRAM requirements. Everything that runs here: what can the RTX 3060 12 GB run?

Other models that run on RTX 3060 12 GB

Questions

Can Medgemma 1.5 4B IT run on RTX 3060 12 GB?

Medgemma 1.5 4B IT at 8-bit runs on the RTX 3060 12 GB with 128K of context at about 61 tokens a second.

Report VR-1EF9JPO, calculated Nov 22, 2026 from published specs, not benchmarked.