Back to search results
Qwen3-4B-Thinking-2507 · Q3_K_M
Apple M2 UltraOllama v0.46L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-4B-Thinking-2507 (Q3_K_M) on Apple M2 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
131.4 tok/s
Decode
Decode speed
507.32 tok/s
Prefill
Prefill speed
1.91 s
TTFT
Time to first token
7.81 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- vLLM · IQ2_M Decode 191.2 · Prefill 838.86 ·
- Ollama · Q3_K_M (current)Decode 131.4 · Prefill 507.32 ·
Core figures
- Decode (avg)
- 131.4 tok/s
- Prefill (avg)
- 507.32 tok/s
- TTFT (avg)
- 1.91 s
- VRAM (avg)
- 7.81 GB
- MTP acceptance rate
- —
- TTFB
- 1.05 GB
- Power draw
- 64.2 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-4B-Thinking-2507
- Quantization
- Q3_K_M
- Framework
- Ollama
- Version
- v0.46
- Context length
- 8192 tokens
- Batch size
- 512
- GPU layers
- 20
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Apple M2 Ultra
- Measured VRAM (avg)
- 7.81 GB
- OS
- macOS 15.0
- Driver
- Apple M2 Ultra
- CUDA
- N/A
- Power draw
- 64.2 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.46 · macOS 15.0 · CUDA N/A · 8192 ctx
131.4 tok/s
Decode
507.32 tok/s
Prefill
1.91 s
TTFT
7.81 GB
VRAM
—
MTP
64.2 W
Power draw
L2 batch seed 2026-10-16: Ollama on hw_id=103 running model_id=57 at Q3_K_M