Back to search results
Qwen3.5-4B · IQ4_XS
Apple M1 MaxStrata v0.31L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3.5-4B (IQ4_XS) on Apple M1 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
158.99 tok/s
Decode
Decode speed
630.94 tok/s
Prefill
Prefill speed
2.16 s
TTFT
Time to first token
8.32 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 57.12 · Prefill 260.15 ·
- vLLM · Q3_K_S Decode 135.63 · Prefill 662.69 ·
- Strata · IQ4_XS (current)Decode 158.99 · Prefill 630.94 ·
Core figures
- Decode (avg)
- 158.99 tok/s
- Prefill (avg)
- 630.94 tok/s
- TTFT (avg)
- 2.16 s
- VRAM (avg)
- 8.32 GB
- MTP acceptance rate
- 0.57
- TTFB
- 1.27 GB
- Power draw
- 42.6 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3.5-4B
- Quantization
- IQ4_XS
- Framework
- Strata
- Version
- v0.31
- Context length
- 2048 tokens
- Batch size
- 256
- GPU layers
- 12
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Apple M1 Max
- Measured VRAM (avg)
- 8.32 GB
- OS
- macOS 15.0
- Driver
- Apple M1 Max
- CUDA
- N/A
- Power draw
- 42.6 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.31 · macOS 15.0 · CUDA N/A · 2048 ctx
158.99 tok/s
Decode
630.94 tok/s
Prefill
2.16 s
TTFT
8.32 GB
VRAM
0.57
MTP
42.6 W
Power draw
L2 batch seed 2026-10-16: Strata on hw_id=101 running model_id=48 at IQ4_XS