Back to search results
Qwen3-8B · Q3_K_S
Apple M2 Ultrallama.cpp v0.53L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen3-8B (Q3_K_S) on Apple M2 Ultra with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
119.1 tok/s
Decode
Decode speed
434.44 tok/s
Prefill
Prefill speed
0.28 s
TTFT
Time to first token
13.02 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 (current)Decode 119.1 · Prefill 434.44 ·
- Ollama · Q2_K Decode 112.25 · Prefill 576.42 ·
- llama.cpp · Q4_K_S Decode 108.67 · Prefill 436.07 ·
- llama.cpp · Q5_K_S Decode 87.67 · Prefill 476.02 ·
- Ollama · Q5_K_M Decode 79.98 · Prefill 403.29 ·
- Ollama · Q8_0 Decode 67.83 · Prefill 284.74 ·
Core figures
- Decode (avg)
- 119.1 tok/s
- Prefill (avg)
- 434.44 tok/s
- TTFT (avg)
- 0.28 s
- VRAM (avg)
- 13.02 GB
- MTP acceptance rate
- —
- TTFB
- 0.95 GB
- Power draw
- 121.4 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen3-8B
- Quantization
- Q3_K_S
- Framework
- llama.cpp
- Version
- v0.53
- Context length
- 16384 tokens
- Batch size
- 256
- GPU layers
- 99
- 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)
- 13.02 GB
- OS
- macOS 26.1
- Power draw
- 121.4 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.53 · macOS 26.1 · 16384 ctx
119.1 tok/s
Decode
434.44 tok/s
Prefill
0.28 s
TTFT
13.02 GB
VRAM
—
MTP
121.4 W
Power draw
L2 batch seed 2026-10-19: llama.cpp on hw_id=103 running model_id=53 at Q3_K_S