Back to search results
Llama-3.1-8B-Instruct · Q3_K_L
Apple M5 Max128GBllama.cpp v0.53L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (Q3_K_L) on Apple M5 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
81.19 tok/s
Decode
Decode speed
358.64 tok/s
Prefill
Prefill speed
1.18 s
TTFT
Time to first token
14.2 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- llama.cpp · IQ2_M Decode 101.46 · Prefill 460.44 ·
- llama.cpp · Q3_K_L (current)Decode 81.19 · Prefill 358.64 ·
- Ollama · IQ3_XS Decode 82.82 · Prefill 392.75 ·
- Ollama · Q3_K_S Decode 79.81 · Prefill 418 ·
Core figures
- Decode (avg)
- 81.19 tok/s
- Prefill (avg)
- 358.64 tok/s
- TTFT (avg)
- 1.18 s
- VRAM (avg)
- 14.2 GB
- MTP acceptance rate
- —
- TTFB
- 0.65 GB
- Power draw
- 60.9 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3.1-8B-Instruct
- Quantization
- Q3_K_L
- Framework
- llama.cpp
- Version
- v0.53
- Context length
- 16384 tokens
- Batch size
- 512
- GPU layers
- 33
- 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)
- 14.2 GB
- OS
- macOS 26.1
- Power draw
- 60.9 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.53 · macOS 26.1 · 16384 ctx
81.19 tok/s
Decode
358.64 tok/s
Prefill
1.18 s
TTFT
14.2 GB
VRAM
—
MTP
60.9 W
Power draw
L2 batch seed 2026-10-19: llama.cpp on hw_id=28 running model_id=68 at Q3_K_L