Arao AI
Back to search results

gemma-4-12B-it · IQ3_XS

Start a discussion
NVIDIA RTX 309024GBOllama v0.49L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-12B-it (IQ3_XS) on NVIDIA RTX 3090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

67.34 tok/s

Decode

Decode speed

278.97 tok/s

Prefill

Prefill speed

2.01 s

TTFT

Time to first token

17.72 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · Q6_K Decode 50.6 · Prefill 208.67 ·
  2. Ollama · IQ3_XS (current)Decode 67.34 · Prefill 278.97 ·
  3. vLLM · Q4_K_M Decode 100.66 · Prefill 612.88 ·
  4. Ollama · Q2_K Decode 115.28 · Prefill 513.9 ·
  5. Ollama · Q4_K_S Decode 84.3 · Prefill 357.49 ·
  6. Strata · IQ2_M Decode 174.12 · Prefill 755.26 ·
  7. vLLM · Q8_0 Decode 61.2 · Prefill 459 ·
  8. vLLM · Q5_K_M Decode 89.4 · Prefill 670.5 ·
  9. llama.cpp · Q8_0 Decode 56.2 · Prefill 421.5 ·
  10. llama.cpp · Q5_K_M Decode 82 · Prefill 615 ·
  11. llama.cpp · Q4_K_M Decode 100 · Prefill 750 ·

Core figures

Decode (avg)
67.34 tok/s
Prefill (avg)
278.97 tok/s
TTFT (avg)
2.01 s
VRAM (avg)
17.72 GB
MTP acceptance rate
—
TTFB
0.54 GB
Power draw
253.1 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-12B-it
Quantization
IQ3_XS
Framework
Ollama
Version
v0.49
Context length
2048 tokens
Batch size
256
GPU layers
33
Flash Attention
Off

Hardware

Nominal and measured figures are shown side by side; whether it runs is the reader's call

GPU
NVIDIA RTX 3090
Nominal VRAM
24 GB
Measured VRAM (avg)
17.72 GB
OS
Windows 11 23H2
Driver
560.94
CUDA
12.6
Power draw
253.1 W

Sources & evidence

1 measured records in total, each traceable to its original source

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.49 · Windows 11 23H2 · CUDA 12.6 · 2048 ctx

    67.34 tok/s

    Decode

    278.97 tok/s

    Prefill

    2.01 s

    TTFT

    17.72 GB

    VRAM

    —

    MTP

    253.1 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=2 running model_id=65 at IQ3_XS