Arao AI
Back to search results

Qwen3.8-27B · Q4_K_M

Start a discussion
NVIDIA H2097GBStrata v0.43L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.8-27B (Q4_K_M) on NVIDIA H20 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

39.14 tok/s

Decode

Decode speed

148.55 tok/s

Prefill

Prefill speed

0.46 s

TTFT

Time to first token

48.59 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. llama.cpp · IQ2_XS Decode 42.1 · Prefill 229.32 ·
  2. Strata · Q2_K Decode 52.83 · Prefill 205.52 ·
  3. Strata · Q4_K_M (current)Decode 39.14 · Prefill 148.55 ·

Core figures

Decode (avg)
39.14 tok/s
Prefill (avg)
148.55 tok/s
TTFT (avg)
0.46 s
VRAM (avg)
48.59 GB
MTP acceptance rate
0.51
TTFB
0.31 GB
Power draw
312.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.8-27B
Quantization
Q4_K_M
Framework
Strata
Version
v0.43
Context length
8192 tokens
Batch size
512
GPU layers
15
Flash Attention
On

Hardware

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

GPU
NVIDIA H20
Nominal VRAM
97 GB
Measured VRAM (avg)
48.59 GB
OS
Windows 11 23H2
Driver
570.86
CUDA
12.6
Power draw
312.6 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 2026-10-19

    v0.43 · Windows 11 23H2 · CUDA 12.6 · 8192 ctx

    39.14 tok/s

    Decode

    148.55 tok/s

    Prefill

    0.46 s

    TTFT

    48.59 GB

    VRAM

    0.51

    MTP

    312.6 W

    Power draw

    L2 batch seed 2026-10-19: Strata on hw_id=56 running model_id=5 at Q4_K_M