Back to search results
Qwen3.8-9B · Q8_0
AMD Strix Halo (Radeon 8060S)vLLM v0.49L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.8-9B (Q8_0) on AMD Strix Halo (Radeon 8060S) with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
54.8 tok/s
Decode
Decode speed
273.05 tok/s
Prefill
Prefill speed
0.54 s
TTFT
Time to first token
23.84 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- vLLM · Q3_K_L Decode 88.34 · Prefill 283.42 ·
- vLLM · Q8_0 (current)Decode 54.8 · Prefill 273.05 ·
- Strata · IQ3_XS Decode 138.39 · Prefill 445.74 ·
- Strata · Q5_K_S Decode 95.58 · Prefill 525.15 ·
Core figures
- Decode (avg)
- 54.8 tok/s
- Prefill (avg)
- 273.05 tok/s
- TTFT (avg)
- 0.54 s
- VRAM (avg)
- 23.84 GB
- MTP acceptance rate
- —
- TTFB
- 0.44 GB
- Power draw
- 94.9 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.8-9B
- Quantization
- Q8_0
- Framework
- vLLM
- Version
- v0.49
- Context length
- 8192 tokens
- Batch size
- 8
- GPU layers
- 999
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- AMD Strix Halo (Radeon 8060S)
- Measured VRAM (avg)
- 23.84 GB
- OS
- Windows 11 23H2
- Driver
- amdgpu 6.11
- CUDA
- N/A
- Power draw
- 94.9 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.49 · Windows 11 23H2 · CUDA N/A · 8192 ctx
54.8 tok/s
Decode
273.05 tok/s
Prefill
0.54 s
TTFT
23.84 GB
VRAM
—
MTP
94.9 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=87 running model_id=52 at Q8_0