Back to search results
Qwen3-4B-Thinking-2507 · Q4_K_M
Apple M1 MaxStrata v0.48L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (Q4_K_M) on Apple M1 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
125.01 tok/s
Decode
Decode speed
495.06 tok/s
Prefill
Prefill speed
1.65 s
TTFT
Time to first token
8.99 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 48.27 · Prefill 248.84 ·
- Strata · IQ3_XS Decode 182.49 · Prefill 683.44 ·
- Strata · Q4_K_M (current)Decode 125.01 · Prefill 495.06 ·
Core figures
- Decode (avg)
- 125.01 tok/s
- Prefill (avg)
- 495.06 tok/s
- TTFT (avg)
- 1.65 s
- VRAM (avg)
- 8.99 GB
- MTP acceptance rate
- 0.57
- TTFB
- 1 GB
- Power draw
- 54.9 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- Q4_K_M
- Framework
- Strata
- Version
- v0.48
- Context length
- 32768 tokens
- Batch size
- 512
- GPU layers
- 5
- Flash Attention
- Off
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)
- 8.99 GB
- OS
- macOS 15.0
- Driver
- Apple M1 Max
- CUDA
- N/A
- Power draw
- 54.9 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.48 · macOS 15.0 · CUDA N/A · 32768 ctx
125.01 tok/s
Decode
495.06 tok/s
Prefill
1.65 s
TTFT
8.99 GB
VRAM
0.57
MTP
54.9 W
Power draw
L2 batch seed 2026-10-16: Strata on hw_id=101 running model_id=57 at Q4_K_M