Back to search results
Qwen3-8B · FP16
Apple M3 Max36GBllama.cpp v0.28L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-8B (FP16) on Apple M3 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
58.36 tok/s
Decode
Decode speed
290.84 tok/s
Prefill
Prefill speed
2.27 s
TTFT
Time to first token
33.18 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
Today
Last verified
Performance
- llama.cpp · FP16 (current)Decode 58.36 · Prefill 290.84 ·
- llama.cpp · IQ2_M Decode 137.92 · Prefill 603.9 ·
- vLLM · FP16 Decode 57.8 · Prefill 255.66 ·
- vLLM · IQ2_M Decode 145.31 · Prefill 613.3 ·
- Ollama · IQ2_M Decode 129.83 · Prefill 636.38 ·
Core figures
- Decode (avg)
- 58.36 tok/s
- Prefill (avg)
- 290.84 tok/s
- TTFT (avg)
- 2.27 s
- VRAM (avg)
- 33.18 GB
- MTP acceptance rate
- —
- TTFB
- 0.47 GB
- Power draw
- 51.7 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-8B
- Quantization
- FP16
- Framework
- llama.cpp
- Version
- v0.28
- Context length
- 16384 tokens
- Batch size
- 512
- GPU layers
- 28
- Flash Attention
- On
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Apple M3 Max
- Nominal VRAM
- 36 GB
- Measured VRAM (avg)
- 33.18 GB
- OS
- macOS 15.0
- Driver
- Apple M3 Max
- CUDA
- N/A
- Power draw
- 51.7 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.28 · macOS 15.0 · CUDA N/A · 16384 ctx
58.36 tok/s
Decode
290.84 tok/s
Prefill
2.27 s
TTFT
33.18 GB
VRAM
—
MTP
51.7 W
Power draw
L2 batch seed 2026-10-10: llama.cpp on hw_id=4 running model_id=53 at FP16