Back to search results
Qwen3.5-9B · IQ3_S
Apple M5 Max128GBvLLM v0.35L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-9B (IQ3_S) on Apple M5 Max with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.
222.48 tok/s
Decode
Decode speed
1,168.2 tok/s
Prefill
Prefill speed
2.41 s
TTFT
Time to first token
14.57 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- vLLM · IQ3_S (current)Decode 222.48 · Prefill 1168.2 ·
- vLLM · Q5_K_S Decode 163.05 · Prefill 700.99 ·
- Strata · Q3_K_M Decode 284.27 · Prefill 948.35 ·
- vLLM · Q5_K_M Decode 148.01 · Prefill 738.05 ·
Core figures
- Decode (avg)
- 222.48 tok/s
- Prefill (avg)
- 1,168.2 tok/s
- TTFT (avg)
- 2.41 s
- VRAM (avg)
- 14.57 GB
- MTP acceptance rate
- —
- TTFB
- 1.78 GB
- Power draw
- 244.4 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-9B
- Quantization
- IQ3_S
- Framework
- vLLM
- Version
- v0.35
- Context length
- 8192 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 M5 Max
- Nominal VRAM
- 128 GB
- Measured VRAM (avg)
- 14.57 GB
- OS
- Ubuntu 24.04
- Driver
- 565.57
- CUDA
- 12.8
- Power draw
- 244.4 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.35 · Ubuntu 24.04 · CUDA 12.8 · 8192 ctx
222.48 tok/s
Decode
1,168.2 tok/s
Prefill
2.41 s
TTFT
14.57 GB
VRAM
—
MTP
244.4 W
Power draw
L2 batch seed 2026-10-16: vLLM on hw_id=28 running model_id=49 at IQ3_S