Arao AI
Back to search results

mammoth-1 · Q4_K_S

Start a discussion
AMD Ryzen AI Max+ 395128GBllama.cpp v0.51L2 Cross-framework verified

This page aggregates 1 real-world runs of mammoth-1 (Q4_K_S) on AMD Ryzen AI Max+ 395 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

26.2 tok/s

Decode

Decode speed

94.34 tok/s

Prefill

Prefill speed

0.65 s

TTFT

Time to first token

46.33 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

13 days ago

Last verified

Performance

  1. llama.cpp · Q4_K_S (current)Decode 26.2 · Prefill 94.34 ·
  2. vLLM · Q4_K_S Decode 27.6 · Prefill 131.32 ·

Core figures

Decode (avg)
26.2 tok/s
Prefill (avg)
94.34 tok/s
TTFT (avg)
0.65 s
VRAM (avg)
46.33 GB
MTP acceptance rate
—
TTFB
0.21 GB
Power draw
74.8 W

Configuration

Member-level fields are taken from the most recent run

Model
mammoth-1
Quantization
Q4_K_S
Framework
llama.cpp
Version
v0.51
Context length
8192 tokens
Batch size
256
GPU layers
99
Flash Attention
On

Hardware

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

GPU
AMD Ryzen AI Max+ 395
Nominal VRAM
128 GB
Measured VRAM (avg)
46.33 GB
OS
Windows 11 23H2
Driver
amdgpu 6.12
CUDA
12.6
Power draw
74.8 W

Get started

Original model resources

Base model resources may not match this quantization. Use the verified deployment weights above when available.

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-22

    v0.51 · Windows 11 23H2 · CUDA 12.6 · 8192 ctx

    26.2 tok/s

    Decode

    94.34 tok/s

    Prefill

    0.65 s

    TTFT

    46.33 GB

    VRAM

    —

    MTP

    74.8 W

    Power draw

    L2 batch seed 2026-10-22: llama.cpp on hw_id=24 running model_id=94 at Q4_K_S