Back to search results
Qwen2.5-Omni-3B · Q6_K
2x AMD Strix HalovLLM v0.37L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (Q6_K) on 2x AMD Strix Halo with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
171.22 tok/s
Decode
Decode speed
756.33 tok/s
Prefill
Prefill speed
1.1 s
TTFT
Time to first token
8.4 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- llama.cpp · IQ2_XS Decode 273.09 · Prefill 1000.97 ·
- llama.cpp · IQ3_S Decode 239.07 · Prefill 794.3 ·
- vLLM · Q5_K_M Decode 183.47 · Prefill 864.96 ·
- vLLM · Q6_K (current)Decode 171.22 · Prefill 756.33 ·
Core figures
- Decode (avg)
- 171.22 tok/s
- Prefill (avg)
- 756.33 tok/s
- TTFT (avg)
- 1.1 s
- VRAM (avg)
- 8.4 GB
- MTP acceptance rate
- —
- TTFB
- 1.37 GB
- Power draw
- 72.1 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-Omni-3B
- Quantization
- Q6_K
- Framework
- vLLM
- Version
- v0.37
- Context length
- 32768 tokens
- Batch size
- 32
- GPU layers
- 999
- 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)
- 8.4 GB
- OS
- Ubuntu 22.04
- Driver
- 560.94
- CUDA
- N/A
- Power draw
- 72.1 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.37 · Ubuntu 22.04 · CUDA N/A · 32768 ctx
171.22 tok/s
Decode
756.33 tok/s
Prefill
1.1 s
TTFT
8.4 GB
VRAM
—
MTP
72.1 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=88 running model_id=60 at Q6_K