Back to search results
Qwen3.5-0.8B · IQ3_XS
Intel Arc B58012GBllama.cpp v0.53L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-0.8B (IQ3_XS) on Intel Arc B580 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
131.95 tok/s
Decode
Decode speed
488.13 tok/s
Prefill
Prefill speed
0.49 s
TTFT
Time to first token
3.14 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
17 days ago
Last verified
Performance
- llama.cpp · IQ3_S Decode 138.57 · Prefill 710.44 ·
- llama.cpp · IQ3_XS (current)Decode 131.95 · Prefill 488.13 ·
- Ollama · Q4_K_S Decode 101.11 · Prefill 350.9 ·
- Ollama · Q8_0 Decode 66.71 · Prefill 216.71 ·
- Ollama · Q3_K_M Decode 110.96 · Prefill 517.6 ·
- Strata · Q5_K_M Decode 136.37 · Prefill 553.15 ·
- llama.cpp · Q4_K_M Decode 117.19 · Prefill 568.87 ·
- vLLM · Q4_K_S Decode 137.77 · Prefill 674.05 ·
Core figures
- Decode (avg)
- 131.95 tok/s
- Prefill (avg)
- 488.13 tok/s
- TTFT (avg)
- 0.49 s
- VRAM (avg)
- 3.14 GB
- MTP acceptance rate
- —
- TTFB
- 1.06 GB
- Power draw
- 144.3 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-0.8B
- Quantization
- IQ3_XS
- Framework
- llama.cpp
- Version
- v0.53
- Context length
- 2048 tokens
- Batch size
- 256
- GPU layers
- 28
- Flash Attention
- Off
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)
- 3.14 GB
- OS
- Windows 11 24H2
- Driver
- Intel Arc 32.0.101.6552
- CUDA
- 12.4
- Power draw
- 144.3 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.53 · Windows 11 24H2 · CUDA 12.4 · 2048 ctx
131.95 tok/s
Decode
488.13 tok/s
Prefill
0.49 s
TTFT
3.14 GB
VRAM
—
MTP
144.3 W
Power draw
L2 batch seed 2026-10-19: llama.cpp on hw_id=94 running model_id=51 at IQ3_XS