Arao AI
Back to search results

Qwen3.5-0.8B · IQ2_M

Start a discussion
NVIDIA RTX 508016GBllama.cpp v0.27L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-0.8B (IQ2_M) on NVIDIA RTX 5080 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

338.41 tok/s

Decode

Decode speed

1,613.66 tok/s

Prefill

Prefill speed

0.9 s

TTFT

Time to first token

2.97 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ2_M (current)Decode 338.41 · Prefill 1613.66 ·
  2. Strata · FP16 Decode 178.36 · Prefill 737.11 ·
  3. Strata · Q4_K_S Decode 363.26 · Prefill 1495.67 ·

Core figures

Decode (avg)
338.41 tok/s
Prefill (avg)
1,613.66 tok/s
TTFT (avg)
0.9 s
VRAM (avg)
2.97 GB
MTP acceptance rate
—
TTFB
2.71 GB
Power draw
255.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-0.8B
Quantization
IQ2_M
Framework
llama.cpp
Version
v0.27
Context length
8192 tokens
Batch size
256
GPU layers
28
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 5080
Nominal VRAM
16 GB
Measured VRAM (avg)
2.97 GB
OS
Ubuntu 22.04
Driver
555.85
CUDA
12.4
Power draw
255.4 W

Sources & evidence

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

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

    v0.27 · Ubuntu 22.04 · CUDA 12.4 · 8192 ctx

    338.41 tok/s

    Decode

    1,613.66 tok/s

    Prefill

    0.9 s

    TTFT

    2.97 GB

    VRAM

    —

    MTP

    255.4 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=14 running model_id=51 at IQ2_M