Back to search results
Qwen3-1.7B · IQ3_XS
Qualcomm Snapdragon 7 Gen 4vLLM v0.49L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-1.7B (IQ3_XS) on Qualcomm Snapdragon 7 Gen 4 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
85.76 tok/s
Decode
Decode speed
337.65 tok/s
Prefill
Prefill speed
0.54 s
TTFT
Time to first token
4.32 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_L Decode 72.64 · Prefill 300.08 ·
- vLLM · IQ3_XS (current)Decode 85.76 · Prefill 337.65 ·
- vLLM · Q3_K_L Decode 76.96 · Prefill 347.46 ·
Core figures
- Decode (avg)
- 85.76 tok/s
- Prefill (avg)
- 337.65 tok/s
- TTFT (avg)
- 0.54 s
- VRAM (avg)
- 4.32 GB
- MTP acceptance rate
- —
- TTFB
- 0.69 GB
- Power draw
- 30.9 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-1.7B
- Quantization
- IQ3_XS
- Framework
- vLLM
- Version
- v0.49
- Context length
- 2048 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
- Qualcomm Snapdragon 7 Gen 4
- Measured VRAM (avg)
- 4.32 GB
- OS
- Windows 11 23H2
- Driver
- Qualcomm Adreno 7.0
- CUDA
- N/A
- Power draw
- 30.9 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.49 · Windows 11 23H2 · CUDA N/A · 2048 ctx
85.76 tok/s
Decode
337.65 tok/s
Prefill
0.54 s
TTFT
4.32 GB
VRAM
—
MTP
30.9 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=105 running model_id=55 at IQ3_XS