Back to search results
Qwen2.5-Omni-3B · Q3_K_M
Apple M1 Maxllama.cpp v0.33L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (Q3_K_M) on Apple M1 Max with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
128.76 tok/s
Decode
Decode speed
591.62 tok/s
Prefill
Prefill speed
1.73 s
TTFT
Time to first token
6.38 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- llama.cpp · FP16 Decode 57.81 · Prefill 312.83 ·
- llama.cpp · Q3_K_M (current)Decode 128.76 · Prefill 591.62 ·
- Ollama · Q4_K_S Decode 116.7 · Prefill 504.6 ·
Core figures
- Decode (avg)
- 128.76 tok/s
- Prefill (avg)
- 591.62 tok/s
- TTFT (avg)
- 1.73 s
- VRAM (avg)
- 6.38 GB
- MTP acceptance rate
- —
- TTFB
- 1.03 GB
- Power draw
- 45.6 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-Omni-3B
- Quantization
- Q3_K_M
- Framework
- llama.cpp
- Version
- v0.33
- Context length
- 32768 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 M1 Max
- Measured VRAM (avg)
- 6.38 GB
- OS
- macOS 15.0
- Driver
- Apple M1 Max
- CUDA
- N/A
- Power draw
- 45.6 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.33 · macOS 15.0 · CUDA N/A · 32768 ctx
128.76 tok/s
Decode
591.62 tok/s
Prefill
1.73 s
TTFT
6.38 GB
VRAM
—
MTP
45.6 W
Power draw
L2 batch seed 2026-10-16: llama.cpp on hw_id=101 running model_id=60 at Q3_K_M