Back to search results
Qwen2.5-Omni-3B · IQ2_XS
Apple M2 Ultrallama.cpp v0.55L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (IQ2_XS) on Apple M2 Ultra with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
215.16 tok/s
Decode
Decode speed
690.54 tok/s
Prefill
Prefill speed
2.04 s
TTFT
Time to first token
5.12 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 · IQ2_XS (current)Decode 215.16 · Prefill 690.54 ·
- llama.cpp · Q3_K_M Decode 171.07 · Prefill 900.09 ·
- vLLM · Q4_K_M Decode 157.8 · Prefill 762.2 ·
Core figures
- Decode (avg)
- 215.16 tok/s
- Prefill (avg)
- 690.54 tok/s
- TTFT (avg)
- 2.04 s
- VRAM (avg)
- 5.12 GB
- MTP acceptance rate
- —
- TTFB
- 1.72 GB
- Power draw
- 37.3 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-Omni-3B
- Quantization
- IQ2_XS
- Framework
- llama.cpp
- Version
- v0.55
- Context length
- 8192 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 M2 Ultra
- Measured VRAM (avg)
- 5.12 GB
- OS
- macOS 15.0
- Driver
- Apple M2 Ultra
- CUDA
- N/A
- Power draw
- 37.3 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.55 · macOS 15.0 · CUDA N/A · 8192 ctx
215.16 tok/s
Decode
690.54 tok/s
Prefill
2.04 s
TTFT
5.12 GB
VRAM
—
MTP
37.3 W
Power draw
L2 batch seed 2026-10-16: llama.cpp on hw_id=103 running model_id=60 at IQ2_XS