T4-BENCH

Measured on Kaggle · 2026

What runs on
two Tesla T4s?

Measured numbers from image, video, language and training runs on the same free notebook. The figures come from our own runs.

2x Tesla T4 15 GB, 31 GB RAM, 4 CPU cores, CUDA 13, PyTorch 2.11, sm_75 Turing.
Fastest chat model
LFM2.5-8B-A1B119 tokens / second
Best small agent
Spark 4B3/3 agent tasks passed
Leanest fine-tune
Unsloth NF47.1 GB graphics memory
Fastest image model
FLUX.2 klein 4B374 images / hour
01

Chat & coding models

How fast each model replies once it is loaded and warm.

Reply speed

Higher is better · tokens per second

Watch the ranking

Diagram from the same data

A token is roughly a word or part of a word.

Two agents at once

Same box, parallel requests

Tool-use test

15 real tasks · score out of 100
02

Agents & long memory

Small models that can use tools and remember long documents.

Agent tasks completed

Out of 3 · higher is better

Ternary 27B reply speed

Tokens per second · higher is better

The larger Spark model finishes more agent tasks, but needs more memory. Adding a second T4 nearly doubles reply speed for the ternary model.

03

Fine-tuning

Memory needed to train an 8B model with the same settings.

Memory used

GB · lower is better

Watch the memory race

Diagram from the same data

Quality was effectively identical between the two working methods, so memory use is the deciding factor.

04

Image generation

Four models, one engine, 1024×1024 pictures.

Pictures per hour

Higher is better · both GPUs busy
05

Video generation

Short clips. Wall time includes loading, generating and saving.

Seconds per clip

Lower is better
06

The memory limit

Why the largest image model needs tricks to run at all.

Memory needed

GB · one T4 has 15 GB
07

How to read this

Every number in these tables comes from a saved run.

What we learned

Small and clever beats big on this machine: a 6B image model outruns a 33B one, and an 8B mixture-of-experts chat model outruns a 27B dense one.

Important limits