Back to search results
Qwen3-4B-Thinking-2507 · IQ2_M
Apple M2 ProStrata v0.32L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (IQ2_M) on Apple M2 Pro with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.
171.6 tok/s
Decode
Decode speed
574.16 tok/s
Prefill
Prefill speed
1.84 s
TTFT
Time to first token
6.47 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 Decode 112.31 · Prefill 603.83 ·
- Strata · IQ2_M (current)Decode 171.6 · Prefill 574.16 ·
- Strata · Q4_K_S Decode 119.23 · Prefill 515.15 ·
Core figures
- Decode (avg)
- 171.6 tok/s
- Prefill (avg)
- 574.16 tok/s
- TTFT (avg)
- 1.84 s
- VRAM (avg)
- 6.47 GB
- MTP acceptance rate
- 0.76
- TTFB
- 1.37 GB
- Power draw
- 51.6 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- IQ2_M
- Framework
- Strata
- Version
- v0.32
- Context length
- 8192 tokens
- Batch size
- 256
- GPU layers
- 8
- 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)
- 6.47 GB
- OS
- macOS 15.0
- Driver
- Apple M2 Pro
- CUDA
- N/A
- Power draw
- 51.6 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.32 · macOS 15.0 · CUDA N/A · 8192 ctx
171.6 tok/s
Decode
574.16 tok/s
Prefill
1.84 s
TTFT
6.47 GB
VRAM
0.76
MTP
51.6 W
Power draw
L2 batch seed 2026-10-16: Strata on hw_id=102 running model_id=57 at IQ2_M