Back to search results
Qwen3-1.7B · Q2_K
Apple M2 ProvLLM v0.55L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-1.7B (Q2_K) on Apple M2 Pro with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
177.75 tok/s
Decode
Decode speed
635.73 tok/s
Prefill
Prefill speed
1.41 s
TTFT
Time to first token
4.06 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 · IQ3_XS Decode 176.88 · Prefill 629.62 ·
- llama.cpp · Q3_K_S Decode 168.44 · Prefill 689.11 ·
- vLLM · Q2_K (current)Decode 177.75 · Prefill 635.73 ·
- vLLM · Q3_K_S Decode 168.62 · Prefill 623.7 ·
Core figures
- Decode (avg)
- 177.75 tok/s
- Prefill (avg)
- 635.73 tok/s
- TTFT (avg)
- 1.41 s
- VRAM (avg)
- 4.06 GB
- MTP acceptance rate
- —
- TTFB
- 1.42 GB
- Power draw
- 70.4 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-1.7B
- Quantization
- Q2_K
- Framework
- vLLM
- Version
- v0.55
- 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
- Apple M2 Pro
- Measured VRAM (avg)
- 4.06 GB
- OS
- macOS 15.0
- Driver
- Apple M2 Pro
- CUDA
- N/A
- Power draw
- 70.4 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.55 · macOS 15.0 · CUDA N/A · 2048 ctx
177.75 tok/s
Decode
635.73 tok/s
Prefill
1.41 s
TTFT
4.06 GB
VRAM
—
MTP
70.4 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=102 running model_id=55 at Q2_K