Back to search results
Qwen3-4B · Q6_K
Apple M2 ProvLLM v0.51L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B (Q6_K) on Apple M2 Pro with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
72.96 tok/s
Decode
Decode speed
295.88 tok/s
Prefill
Prefill speed
2.34 s
TTFT
Time to first token
10.5 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- vLLM · FP16 Decode 48.46 · Prefill 199.81 ·
- vLLM · Q6_K (current)Decode 72.96 · Prefill 295.88 ·
- Strata · FP16 Decode 65.24 · Prefill 265.48 ·
Core figures
- Decode (avg)
- 72.96 tok/s
- Prefill (avg)
- 295.88 tok/s
- TTFT (avg)
- 2.34 s
- VRAM (avg)
- 10.5 GB
- MTP acceptance rate
- —
- TTFB
- 0.58 GB
- Power draw
- 23.2 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B
- Quantization
- Q6_K
- Framework
- vLLM
- Version
- v0.51
- Context length
- 16384 tokens
- Batch size
- 32
- 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)
- 10.5 GB
- OS
- macOS 15.0
- Driver
- Apple M2 Pro
- CUDA
- N/A
- Power draw
- 23.2 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.51 · macOS 15.0 · CUDA N/A · 16384 ctx
72.96 tok/s
Decode
295.88 tok/s
Prefill
2.34 s
TTFT
10.5 GB
VRAM
—
MTP
23.2 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=102 running model_id=54 at Q6_K