Back to search results
Qwen3.5-9B · IQ3_S
AMD Strix Point (Radeon 890M)vLLM v0.21L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-9B (IQ3_S) on AMD Strix Point (Radeon 890M) with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
65.94 tok/s
Decode
Decode speed
339.59 tok/s
Prefill
Prefill speed
2.5 s
TTFT
Time to first token
14.57 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- vLLM · IQ3_S (current)Decode 65.94 · Prefill 339.59 ·
- Ollama · IQ3_XS Decode 62.85 · Prefill 248.56 ·
- Ollama · IQ4_XS Decode 46.77 · Prefill 176.41 ·
Core figures
- Decode (avg)
- 65.94 tok/s
- Prefill (avg)
- 339.59 tok/s
- TTFT (avg)
- 2.5 s
- VRAM (avg)
- 14.57 GB
- MTP acceptance rate
- —
- TTFB
- 0.53 GB
- Power draw
- 86 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-9B
- Quantization
- IQ3_S
- Framework
- vLLM
- Version
- v0.21
- Context length
- 32768 tokens
- Batch size
- 16
- 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 Point (Radeon 890M)
- Measured VRAM (avg)
- 14.57 GB
- OS
- Ubuntu 22.04
- Driver
- 555.85
- CUDA
- N/A
- Power draw
- 86 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.21 · Ubuntu 22.04 · CUDA N/A · 32768 ctx
65.94 tok/s
Decode
339.59 tok/s
Prefill
2.5 s
TTFT
14.57 GB
VRAM
—
MTP
86 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=89 running model_id=49 at IQ3_S