Arao AI
Back to search results

Qwen3.5-27B · Q4_K_M

Start a discussion
NVIDIA RTX 507012GBStrata v0.5.xL2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-27B (Q4_K_M) on NVIDIA RTX 5070 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

64 tok/s

Decode

Decode speed

768 tok/s

Prefill

Prefill speed

1.05 s

TTFT

Time to first token

— GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · IQ4_XS Decode 54 · Prefill 640 ·
  2. llama.cpp · Q4_K_M Decode 50 · Prefill 586 ·
  3. llama.cpp · Q5_K_M Decode 52 · Prefill 454 ·
  4. llama.cpp · Q8_0 Decode 36 · Prefill 394 ·
  5. vLLM · IQ4_XS Decode 51 · Prefill 486 ·
  6. vLLM · Q4_K_M Decode 47 · Prefill 429 ·
  7. vLLM · Q5_K_M Decode 42 · Prefill 447 ·
  8. vLLM · Q8_0 Decode 28 · Prefill 334 ·
  9. Ollama · IQ4_XS Decode 55 · Prefill 549 ·
  10. Ollama · Q4_K_M Decode 50 · Prefill 516 ·
  11. Ollama · Q5_K_M Decode 42 · Prefill 504 ·
  12. Ollama · Q8_0 Decode 36 · Prefill 356 ·
  13. Strata · IQ4_XS Decode 70 · Prefill 837 ·
  14. Strata · Q4_K_M (current)Decode 64 · Prefill 768 ·
  15. Strata · Q5_K_M Decode 72 · Prefill 739 ·
  16. Strata · Q8_0 Decode 53 · Prefill 509 ·

Core figures

Decode (avg)
64 tok/s
Prefill (avg)
768 tok/s
TTFT (avg)
1.05 s
VRAM (avg)
— GB
MTP acceptance rate
—
TTFB
— GB
Power draw
489 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-27B
Quantization
Q4_K_M
Framework
Strata
Version
v0.5.x
Context length
8192 tokens
Batch size
512
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
NVIDIA RTX 5070
Nominal VRAM
12 GB
OS
Linux 6.8.0
Driver
570.86.15
CUDA
12.6
Power draw
489 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-09

    v0.5.x · Linux 6.8.0 · CUDA 12.6 · 8192 ctx

    64 tok/s

    Decode

    768 tok/s

    Prefill

    1.05 s

    TTFT

    0 GB

    VRAM

    —

    MTP

    489 W

    Power draw

    Auto-generated LLM-sourced record 2026-10-09. Model: Qwen3.5-27B, Hardware: RTX 5070, Framework: 9, Quant: Q4_K_M.