Back to search results
Qwen3-4B-Thinking-2507 · Q2_K
Qualcomm Snapdragon X Elitellama.cpp v0.23L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (Q2_K) on Qualcomm Snapdragon X Elite with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
81.63 tok/s
Decode
Decode speed
398.71 tok/s
Prefill
Prefill speed
0.96 s
TTFT
Time to first token
6.72 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 · Q2_K (current)Decode 81.63 · Prefill 398.71 ·
- Strata · Q8_0 Decode 48.7 · Prefill 212.72 ·
- llama.cpp · Q8_0 Decode 18.5 · Prefill 2.2 ·
Core figures
- Decode (avg)
- 81.63 tok/s
- Prefill (avg)
- 398.71 tok/s
- TTFT (avg)
- 0.96 s
- VRAM (avg)
- 6.72 GB
- MTP acceptance rate
- —
- TTFB
- 0.65 GB
- Power draw
- 34.1 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- Q2_K
- Framework
- llama.cpp
- Version
- v0.23
- Context length
- 2048 tokens
- Batch size
- 256
- GPU layers
- 20
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Qualcomm Snapdragon X Elite
- Measured VRAM (avg)
- 6.72 GB
- OS
- Windows 11 23H2
- Driver
- Qualcomm Adreno 7.0
- CUDA
- N/A
- Power draw
- 34.1 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.23 · Windows 11 23H2 · CUDA N/A · 2048 ctx
81.63 tok/s
Decode
398.71 tok/s
Prefill
0.96 s
TTFT
6.72 GB
VRAM
—
MTP
34.1 W
Power draw
L2 batch seed 2026-10-16: llama.cpp on hw_id=104 running model_id=57 at Q2_K