Back to search results
Qwen2.5-Omni-3B · Q4_K_S
Apple M1 MaxOllama v0.24L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (Q4_K_S) on Apple M1 Max with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
116.7 tok/s
Decode
Decode speed
504.6 tok/s
Prefill
Prefill speed
2.46 s
TTFT
Time to first token
7.01 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 Decode 128.76 · Prefill 591.62 ·
- Ollama · Q4_K_S (current)Decode 116.7 · Prefill 504.6 ·
Core figures
- Decode (avg)
- 116.7 tok/s
- Prefill (avg)
- 504.6 tok/s
- TTFT (avg)
- 2.46 s
- VRAM (avg)
- 7.01 GB
- MTP acceptance rate
- —
- TTFB
- 0.93 GB
- Power draw
- 32.5 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-Omni-3B
- Quantization
- Q4_K_S
- Framework
- Ollama
- Version
- v0.24
- Context length
- 4096 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 M1 Max
- Measured VRAM (avg)
- 7.01 GB
- OS
- macOS 15.0
- Driver
- Apple M1 Max
- CUDA
- N/A
- Power draw
- 32.5 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.24 · macOS 15.0 · CUDA N/A · 4096 ctx
116.7 tok/s
Decode
504.6 tok/s
Prefill
2.46 s
TTFT
7.01 GB
VRAM
—
MTP
32.5 W
Power draw
L2 batch seed 2026-10-16: Ollama on hw_id=101 running model_id=60 at Q4_K_S