Back to search results
Qwen2.5-Omni-3B · Q5_K_M
Qualcomm Snapdragon 7 Gen 4Strata v0.31L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (Q5_K_M) on Qualcomm Snapdragon 7 Gen 4 with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
64.91 tok/s
Decode
Decode speed
252.46 tok/s
Prefill
Prefill speed
0.48 s
TTFT
Time to first token
7.6 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 · Q3_K_M Decode 59.3 · Prefill 285.77 ·
- Strata · Q3_K_S Decode 97.85 · Prefill 474.54 ·
- Strata · Q5_K_M (current)Decode 64.91 · Prefill 252.46 ·
Core figures
- Decode (avg)
- 64.91 tok/s
- Prefill (avg)
- 252.46 tok/s
- TTFT (avg)
- 0.48 s
- VRAM (avg)
- 7.6 GB
- MTP acceptance rate
- 0.63
- TTFB
- 0.52 GB
- Power draw
- 26.6 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-Omni-3B
- Quantization
- Q5_K_M
- Framework
- Strata
- Version
- v0.31
- Context length
- 8192 tokens
- Batch size
- 1024
- GPU layers
- 5
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Qualcomm Snapdragon 7 Gen 4
- Measured VRAM (avg)
- 7.6 GB
- OS
- Windows 11 23H2
- Driver
- Qualcomm Adreno 7.0
- CUDA
- N/A
- Power draw
- 26.6 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.31 · Windows 11 23H2 · CUDA N/A · 8192 ctx
64.91 tok/s
Decode
252.46 tok/s
Prefill
0.48 s
TTFT
7.6 GB
VRAM
0.63
MTP
26.6 W
Power draw
L2 batch seed 2026-10-16: Strata on hw_id=105 running model_id=60 at Q5_K_M