ENPL– Polski Check your hardware
All cards

In the measurement queue

NVIDIA GeForce RTX 4070

12 GB · Ada sm_89

We have not measured this card yet.

It is in the queue. Until then we show its spec and an estimate from our formula, clearly labelled.

An estimate, not a measurement

≈38–46t/s

before, estimate · Gemma 4 12B UD-Q4_K_XL · after: soon

Memory
12GB
Bandwidth
504GB/s
vendor spec
Architecture
Ada sm_89
An estimate, not a measurement

What fits

The formula η · bandwidth / bytes per token, for stock llama.cpp. Once measured, real numbers replace it here.

The formula η · bandwidth / bytes per token, for stock llama.cpp. Once measured, real numbers replace it here. app/landing/src/lib/estimator.ts: decodeRange (η·BW/B, raport 04 §2.2.4), memoryNeed (raport 04 §2.2.4)app/landing/src/lib/data/gpus.jsonapp/landing/src/lib/data/models.json
  • Gemma 4 12B UD-Q4_K_XL

    memory needed, estimate ≈ 82–85 % of the card's memory

    before, estimate ≈ 38–46 t/s

    after soon

What next

Check your hardware All cards

Measured cards