Back to search results
Llama-3.2-3B-Instruct · Q3_K_S
Apple M5 Ultra512GBOllama v0.48L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (Q3_K_S) on Apple M5 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
318.38 tok/s
Decode
Decode speed
1,403.79 tok/s
Prefill
Prefill speed
2.49 s
TTFT
Time to first token
6.2 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
17 days ago
Last verified
Performance
- llama.cpp · Q3_K_S Decode 338.26 · Prefill 1583.22 ·
- Ollama · Q3_K_S (current)Decode 318.38 · Prefill 1403.79 ·
Core figures
- Decode (avg)
- 318.38 tok/s
- Prefill (avg)
- 1,403.79 tok/s
- TTFT (avg)
- 2.49 s
- VRAM (avg)
- 6.2 GB
- MTP acceptance rate
- —
- TTFB
- 2.55 GB
- Power draw
- 122 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3.2-3B-Instruct
- Quantization
- Q3_K_S
- Framework
- Ollama
- Version
- v0.48
- Context length
- 4096 tokens
- Batch size
- 512
- GPU layers
- 33
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Apple M5 Ultra
- Nominal VRAM
- 512 GB
- Measured VRAM (avg)
- 6.2 GB
- OS
- macOS 26.1
- Power draw
- 122 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.48 · macOS 26.1 · 4096 ctx
318.38 tok/s
Decode
1,403.79 tok/s
Prefill
2.49 s
TTFT
6.2 GB
VRAM
—
MTP
122 W
Power draw
L2 batch seed 2026-10-19: Ollama on hw_id=29 running model_id=70 at Q3_K_S