Back to search results
Qwen3.5-4B · Q3_K_L
Intel Arc B58012GBllama.cpp v0.45L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-4B (Q3_K_L) on Intel Arc B580 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
70.52 tok/s
Decode
Decode speed
317.93 tok/s
Prefill
Prefill speed
0.3 s
TTFT
Time to first token
8.15 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
- llama.cpp · IQ1_S Decode 98.24 · Prefill 506.66 ·
- llama.cpp · Q3_K_L (current)Decode 70.52 · Prefill 317.93 ·
- vLLM · Q2_K Decode 89.29 · Prefill 348.31 ·
- Strata · IQ3_S Decode 101.32 · Prefill 490.59 ·
Core figures
- Decode (avg)
- 70.52 tok/s
- Prefill (avg)
- 317.93 tok/s
- TTFT (avg)
- 0.3 s
- VRAM (avg)
- 8.15 GB
- MTP acceptance rate
- —
- TTFB
- 0.56 GB
- Power draw
- 267.1 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-4B
- Quantization
- Q3_K_L
- Framework
- llama.cpp
- Version
- v0.45
- Context length
- 32768 tokens
- Batch size
- 512
- GPU layers
- 33
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Intel Arc B580
- Nominal VRAM
- 12 GB
- Measured VRAM (avg)
- 8.15 GB
- OS
- Ubuntu 24.04
- Driver
- Intel Arc 31.0.101.5186
- CUDA
- 12.6
- Power draw
- 267.1 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.45 · Ubuntu 24.04 · CUDA 12.6 · 32768 ctx
70.52 tok/s
Decode
317.93 tok/s
Prefill
0.3 s
TTFT
8.15 GB
VRAM
—
MTP
267.1 W
Power draw
L2 batch seed 2026-10-10: llama.cpp on hw_id=94 running model_id=48 at Q3_K_L