Back to search results
Qwen2.5-0.5B-Instruct · Q8_0
2x AMD Strix HaloOllama v0.51L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-0.5B-Instruct (Q8_0) on 2x AMD Strix Halo with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
256.99 tok/s
Decode
Decode speed
1,220.11 tok/s
Prefill
Prefill speed
0.82 s
TTFT
Time to first token
3.31 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- Ollama · Q3_K_S Decode 347.47 · Prefill 1631.19 ·
- Ollama · Q8_0 (current)Decode 256.99 · Prefill 1220.11 ·
- Strata · Q4_K_M Decode 303.04 · Prefill 1091.32 ·
Core figures
- Decode (avg)
- 256.99 tok/s
- Prefill (avg)
- 1,220.11 tok/s
- TTFT (avg)
- 0.82 s
- VRAM (avg)
- 3.31 GB
- MTP acceptance rate
- —
- TTFB
- 2.06 GB
- Power draw
- 76.3 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-0.5B-Instruct
- Quantization
- Q8_0
- Framework
- Ollama
- Version
- v0.51
- Context length
- 4096 tokens
- Batch size
- 256
- GPU layers
- 99
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- 2x AMD Strix Halo
- Measured VRAM (avg)
- 3.31 GB
- OS
- Windows 11 23H2
- Driver
- 565.57
- CUDA
- N/A
- Power draw
- 76.3 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.51 · Windows 11 23H2 · CUDA N/A · 4096 ctx
256.99 tok/s
Decode
1,220.11 tok/s
Prefill
0.82 s
TTFT
3.31 GB
VRAM
—
MTP
76.3 W
Power draw
L2 batch seed 2026-10-16: Ollama on hw_id=88 running model_id=59 at Q8_0