Back to search results
Qwen2.5-Omni-3B · Q8_0
AMD Ryzen AI Max+ 395128GBllama.cpp v0.29L2 Cross-framework verified
This page aggregates 1 real-world runs of Qwen2.5-Omni-3B (Q8_0) on AMD Ryzen AI Max+ 395 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
98.01 tok/s
Decode
Decode speed
478.15 tok/s
Prefill
Prefill speed
2.36 s
TTFT
Time to first token
9.35 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
5 days ago
Last verified
Performance
- vLLM · IQ3_S Decode 201.15 · Prefill 1047.45 ·
- vLLM · Q2_K Decode 204.49 · Prefill 918.23 ·
- Strata · Q4_K_M Decode 218.64 · Prefill 754.88 ·
- Strata · Q8_0 Decode 147.88 · Prefill 774.09 ·
- llama.cpp · Q4_K_S Decode 153.58 · Prefill 827.06 ·
- llama.cpp · Q8_0 (current)Decode 98.01 · Prefill 478.15 ·
- Ollama · Q3_K_S Decode 173.92 · Prefill 636.41 ·
- Ollama · Q4_K_S Decode 150.76 · Prefill 704.53 ·
Core figures
- Decode (avg)
- 98.01 tok/s
- Prefill (avg)
- 478.15 tok/s
- TTFT (avg)
- 2.36 s
- VRAM (avg)
- 9.35 GB
- MTP acceptance rate
- —
- TTFB
- 0.78 GB
- Power draw
- 105.7 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Qwen2.5-Omni-3B
- Quantization
- Q8_0
- Framework
- llama.cpp
- Version
- v0.29
- Context length
- 2048 tokens
- Batch size
- 512
- GPU layers
- 20
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- AMD Ryzen AI Max+ 395
- Nominal VRAM
- 128 GB
- Measured VRAM (avg)
- 9.35 GB
- OS
- Windows 11 23H2
- Driver
- AMD Adrenalin 25.10.1
- CUDA
- N/A
- Power draw
- 105.7 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.29 · Windows 11 23H2 · CUDA N/A · 2048 ctx
98.01 tok/s
Decode
478.15 tok/s
Prefill
2.36 s
TTFT
9.35 GB
VRAM
—
MTP
105.7 W
Power draw
L2 batch seed 2026-10-13: llama.cpp on hw_id=24 running model_id=60 at Q8_0