Back to search results
Qwen3-4B-Thinking-2507 · IQ4_XS
Apple M5 Max128GBStrata v0.33L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (IQ4_XS) on Apple M5 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
348.81 tok/s
Decode
Decode speed
1,453.67 tok/s
Prefill
Prefill speed
1.79 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
5 days ago
Last verified
Performance
- Ollama · Q5_K_S Decode 192.3 · Prefill 1031.35 ·
- Strata · Q5_K_M Decode 281.24 · Prefill 1191.19 ·
- vLLM · Q4_K_S Decode 240.97 · Prefill 1176.03 ·
- Strata · IQ4_XS (current)Decode 348.81 · Prefill 1453.67 ·
Core figures
- Decode (avg)
- 348.81 tok/s
- Prefill (avg)
- 1,453.67 tok/s
- TTFT (avg)
- 1.79 s
- VRAM (avg)
- 8.32 GB
- MTP acceptance rate
- 0.59
- TTFB
- 2.79 GB
- Power draw
- 58 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- IQ4_XS
- Framework
- Strata
- Version
- v0.33
- Context length
- 2048 tokens
- Batch size
- 512
- GPU layers
- 5
- 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)
- 8.32 GB
- OS
- macOS 15.0
- Driver
- Apple M5 Max
- CUDA
- N/A
- Power draw
- 58 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.33 · macOS 15.0 · CUDA N/A · 2048 ctx
348.81 tok/s
Decode
1,453.67 tok/s
Prefill
1.79 s
TTFT
8.32 GB
VRAM
0.59
MTP
58 W
Power draw
L2 batch seed 2026-10-13: Strata on hw_id=28 running model_id=57 at IQ4_XS